Sonoran Desert

The Sonoran Desert is one of four traditionally recognized deserts in North America, the others in this traditional view being the Chihuhuan, Great Basin, and Mojave Deserts. Some recent (and perhaps more popularized than scientifically based) interpretations added the Colorado Desert as a fifth desert whereas the older perspective regarded the Colorado as one of seven parts (designated as subdivisions by Shreves; explained in detail below) of the greater Sonoran Desert. Broadly speaking the Sonoran and Chihuhuan Deserts are "hot" deserts with their southern portions extending below 30 degrees North latitude into the subtropical belts of high pressure (ie. the horse latitudes). The Great Basin Desert coincides closely (not exactly of course) with the Great Basin section of the Basin and Range province, from which it's designation was derived. The Great Basin Desert is a "cold" desert. The Mojave Desert is the smallest of the traditional four North American deserts and is situated between the "hot" Sonoran Desert and the "cold" Great Basin and has often been characterized as an "intermediate" desert. Brown (1994, ps.143, 144, 156, 180) divided North American deserts ("desertlands" was the overall or general heading) as follows: 1) cold-temperate desertlands- Great Basin desertscrub, 2) warm-temperate desertlands- Mohave desertscrub and Chihuhuan desertlscrub, 3) tropical-subtropical desertlands- Sonoran desertscrub.

All of the North American deserts occur within the vast Basin and Range physiographic province, and all of them have the telltale physiography of valleys or geographic depressions called "basins" situated between mountain ranges of varying elevations but which typically run parallel to the valleys. Ergo: "basin and range" topography. Most of the vast region of the North American deserts was designated as Desert climate (BW) by the Koppen-Geiger system of climate classification although parts of the perimeters of these deserts was classified as Steppe climate (BS). Obviously there is no exact line between these. Big sagebrush-dominated Great Basin Desert blends into the bunchgrass-sagebrush shrub steppe and the creosotebush-dominated Chihuhuan Desert merges with the semidesert grassland. Boundaries of the Sonoran Desert are for the most part more distinct than those of the former two large deserts.

Good summary discussions of the Koppen-Geiger climate classification system are available in most of the standard texts on Climatology or Meteorology including the comprehensive Strahler and Strahler (1978, ps. 144-146). For details of the Basin and Range physiographic province the definitive authority remains the description and maps of Fenenman (1931, Chapter VIII, esp. ps. 367-379 for Sonoran Desert and Salton Basin sections). The more recent text of Hunt (1974, Chapter 16, esp. ps. 499-502 for Sonoran Desert and Salton Trough) is also worth reading as is the work of Laity in Orme (2002, Chapter 19, ps. 390-393 for Sonoran Desert). Readers should avoid confusion over the difference between the Sonoran Desert and it's six vegetational subdivisions and the Sonoran Desert section of the Basin and Range physiographic province. The latter is smaller in geographic area and refers to geological aspects like landforms. The former is essentially a large biological community whose boundaries are functions of climate, landforms, soils, and other abiotic factors along with flora and fauna that comprise the Sonoran Desert ecosystem.

In this publication, which is overwhelmingly Clementsian in perspective and terminology, the Sonoran Desert is one desert that makes up part of the desert biome or, if but for the moment emphasis is laid on vegetation, the desert formation.

For all things Sonoran the "alpha and omega" (at least the beginning) is the life-long work of Forrest Shreve with that of Shreve in Shreve and Wiggins (1964, ps. 6-186 plus plates) the defining ecological monograph. A more recent comprehensive source (and one that still used the seven vegetational subdivisions of Shreve) is Crosswhite and Crosswhite (Chapter 5 in Bender, 1982). General descriptions and discussions of the Sonoran Desert (and also the Mojave and Great Basin Deserts) vegetation can be found in the ecological monographs of North American biotic communities (Shelford, 1963, ps. 373-387) and North American vegetation (Barbour and Billings, 1988; Barbour and Billings 2000)) and that of Californian vegetation (Barbour and Major, 1995). Unfortunately these more recent descripitions ignored the herbaceous layer of vegetation, in particular giving little or no consideration of grasses. Jaeger (1957) has been a choice starting place for general and less-advanced reading on the deserts of North America.

"In a league by itself" except for Shreve is Discovering the Desert by McGinnies (1981). McGinnies was a colleague and contemporary of Shreve and likely had more first-hand knowledge of the Sonoran Desert than any authority except for Shreve. Like Shreve and Clements, McGinnies included grasses and the herbaceous component of the vegetation and did not ignore them as did many later ecologists.

Chapter from history of Ecology: It is of both historical and biological importance to understand the "ecological perspective" of Forrest Shreve because he remains the ecologist who knew the most about the vegetation of the Sonoran Desert. Shreve laid the foundation on which most subsequent and substative ecological studies into the Sonoran Desert have been based (eg. all major contemporary descriptions of the Sonoran Desert retained the seven vegetational subdivisions of Shreve). Forest Shreve was to deserts what John Weaver was to grasslands. Shreve also worked in the other four major deserts of North America, but World War II and his deteriorating health intervened so that Shreve did not achieve "repeat successes" in these communities (Spoehr in Shreve and Wiggins, 1964, p. 5).

Like the better known, more prolific publication-producing, politically astute, and superior self-promoting Frederic Clements and famous "cactophiles" Nathaniel Britton and J.N. Rose, Forrest Shreve worked for the famous Carnegie Institution of Washington. In fact, at one time or the other all of these famous pioneer plant scientists and ecologists worked at the Carnegie Desert Botanical Laboratory near Tucson, Arizona. The fascinating story (to plant ecologists anyway) of Shreve and the intriguing politics (by academic standards) of the Desert Laboratory and related units like the Alpine Laboratory on Pikes Peak, where Clements served in the summer (and the Desert Laboratory in winter) for eight years, was told in factually detailed and lively fashion by Bowers (1988). Bowers' scholarship should be especially interesting and revealing to those versed in the conflict between the Clementsian succession-and-climax model of "dynamic vegetation" and the Gleasonian continuum or individualistic model of vegetation.

[In terms of institutional or office politics and flattering the one who foots the bill, your current author simply drew attention to the scientific name of the plant that to most serves as the iconic personification of the Sonoran Desert: Carnegiea gigantea."Carnegiea is the best know of the characteristic plants of the Sonoran Desert" (Shreve in Shreve and Wiggins, 1964, p. 147).]

Of relevance to this introduction of the Sonoran Desert (and the ecological study thereof) is the fact that Shreve-- who became synonymous with Sonoran Desert research-- was more in the camp of Henry Gleason and the individualistic view of plant communities than in the gathering of orthodox Clementsian disciples. Again the research of Bowers (1988) was to the point and she quoted from pages 21-22 of Shreves culminating publication, Vegetation of the Sonoran Desert (Shreve [1951] as Carnegie Institution of Washington Publication No. 591 and later--and more accessible-- reprinted as pages 1-186 plus plates in Shreve and Wiggins [1964]):

"The successional changes which are so important elsewhere may be read into the vegetation of the desert, but the evidence for them would hardly suggest that they are of importance in determining the relations of the communities to one another. The secular changes of surface are accompanied by the shifting of the communities over long periods, by which their locations and areas ae changed in the simplest possible manner. If a particular community is destroyed without change in the soil, the earliest stage in the return of vegetation will be appearance of young plants of the former dominants. No only do the same species reappear at the outset, but their first individuals ultimately constitute the restored community.

"It is not possible to use the term 'climax' with reference to desert vegetation. Each habitat in each subdivision of a desert area has its own climax, which must be given an elastic definition and must not be interpreted as having a genetic relation to any other climax. It is merely the particular group of species which, in somewhat definite proportions and with a fairly definite communal arrangement, is able to occupy a particular location under its present environmental conditions". (Shreve in Shreve and Wiggins, 1964, ps. 29-30).

The strains of the individualistic concept of Gleason were clear in that passage (last sentence in quotation). So also, however, was the theme of Arthur Tansley's polyclimax concept and, later, of Whittaker's climax pattern theory both of which, as described in the literature review herein, were modifications or amendments of the prevailing Clementsian model of plant succession, only one part of which was the monoclimax view. There is nothing in this explanation of vegetation or desert plant communities by Shreve that placed him in either the Gleason or Clements school. In fact, the second paragraph quoted was consistent with the range site concept which was an outgrowth of the polyclimax perspective of Tansley and of plant indicators tracing to Clements second Carnegie Institution "Sears & Sawbuck"-sized monograph (Clements, 1920).

Conversely, the first paragaraph of Shreve's writing was beyond doubt the continuum interpretation in Gleason's individualistic theory of vegetation. Bowers (1988, p. 50-51) quoted Shreve discussing "environment gradients" and she explained how Robert Whittaker later used this concept in development of gradient anaalysis. She remarked that Shreve had described Jamaican mountain vegetation as a continuum, but that Gleason received credit for the concept though his publication came out two year after the one by Shreve (Bowers, 1988, p. 54). Bowers also specified that "[g]radients were a familar concept, though". Clearly, Shreve was an individualist (in general, not ecological school-wise) who had ideas of his own, most of which remain the foundation for description of the Sonoran Desert.

Bowers (1988, p. 88) concluded that "Shreve's denial of desert succession may have been largely semantic". Bowers illustrated how some of the "processes of vegetational change" Shreve recognized were regarded as plant succession decades later. "If Shreve could have used these looser definitions, he might well have been willing to admit the existence of succession in desert environments" (Bowers, 1988, p. 88). On the other hand Bowers noted that Shreve periodically got in a few jabs at Clements (eg. Bowers, 1988, ps.33, 55) and Shreves "may have felt constrained" "... by the fact that the Carnegie Institution supported them both" (Bowers, 1988, p. 60).

Observe from the above quotation in Vegetation of the Sonoran Desert (Shreve and Wiggins, 1964, ps. 29-30) that Shreve did accept the concept of dominant species, a perspective central to the cover type concept. This concept in turn was the basis of dominance type, the form or category of type used by the Society of American Foresters and Society for Range Management for forest and rangeland cover types, respectively.

The indisuptable fact remains, however, that Shreve opposed many of Clements' central tenets (dogma in the opinions of some). Shreve certainly saw vegetation as a continuum and not as discrete, repeatable, natural units that were the mature or adult "organism" toward which all vegetation in a given climate was invariably developing. Bowers (1988, p. 61) quoted Shreve (1915) as having written that it was "... nowhere possible to pick out a group of plants which may be thought of as associates without being able to find other localitites in which the association has been dissolved". Shreve's Carnegie Institution Publication No. 217, The Vegetation of a Desert Mountain Range as Conditioned by Climatic Factors, predated the monumentally influencial Plant Succession, Carnegie Instution Publication No. 242 (Clements, 1916a) by one year. On no other point central to plant communities did Shreve and Clements have more disparate views. Clements saw discrete, natural plant communities that developed into mature organism-like assemblages of species all of which existed in a hierarchial arrangement of vegetational units; Shreve saw only individual plant communities arrayed along a continnuum of environmental gradients.

Bowers (1988, p. 135) quoted a comment by Shreve to a correspondent, "I take very little interest in the formal treatment of vegetation and the customary descriptions of successions and climaxes treated with great attention to the formal and highly artificial clasifications of which we have so many". Clements, by stark contrast, seemed to thrive on such "formal treatment of vegetation" and he certainly did benefit immeasurably from these which he "packaged" and "marketed" so successfully.

One of Clements' hierarchial units of vegetation that was most influencial with and widely used by applied ecologists like foresters and rangemen was the association. The association was central to the Clementsian concepts of climax, formation, and biome as the mature organism of vegetation and biotic community (see review of literature herein). Each climax formation had two or more major subdivisions designated as associations, each of which was defined or designated by it's one or more dominants (Weaver and Clements, 1938, p. 93).

Shreve (in Shreve and Wiggins, 1964, p. 29) accepted the idea of dominants, but he rejected associations (Shreve, 1915). It was shown in the literature reveiw herein that forest and rangeland cover types are a form of the dominance type that was derived from the association, a unit central to the English or Anglo-American Traditon or School of Plant Ecology (Shimwell, p. 53-54). Historically the association, basically the Clementsian association, and the dominance type were essentially synonymous in American and British plant ecology. (Among the various traditions or schools of Plant Ecology meaning of dominance types varied as, for instance, between the Anglo-American and the Braun-Blanquet school [Whittaker, 1980, ps.67-79].)

Without the association (without the validity of the association) the Anglo-American dominance type would be of questionable meaning or legitimacy. In other words, if the view of Shreve (1915) was correct it would be meaningless (or nearly so) to speak of cover types. This was in net effect the logical extension and practical conclusion of the Shreve-Gleason concept of plant communities as a continuum of individual groupings of plant species arranged along environmental gradients. Precisely speaking Gleason (1917, 1926) did accept plant associations but as individual assemblages, "the individualistic concept of the plant association". Of necessity (or perhaps by default) this concept, now widely endorsed by many plant ecologists, was unaccepted (or ignored) when more-or-less discrete, repeatable cover types were named, described, and published as "standards" of vegetation. Even more, there was a tendency for cover types to be climax vegetation (or disclimaxes in cases of stable communities made up of naturalized species), disclaimers of published cover types notwithstanding. It seems that Shreve lost out to Clements who so "excelled at packaging his wares" (Bowers, 1988, p. 59).

Shreve did "count coup" (sort of ) on Clements in one important and lasting regard. Shreve's vegetational subdivision of the Sonoran Desert remains the one most commonly accepted and widely used arrangement by those working therein. Shreve (in Shreve and Wiggins, 1964, p. 47- 126) based his subdivision or conceptual framework of the Sonoran Desert on three criteria: 1) geographic and physiographic features, 2) abundance of plant genera or species in the distinct geographic-physiographic areas (called "regions"), and 3) "character and organization of the communities of plants". Shreve (in Shreve and Wiggins, 1964, p. 48) presented these as "three sets of equivalent designations for the seven subdivisions" in a format with the last criteria, 3) "character ... of plants", numbered and at top of a "set" with the subdivision itself, 1) "geographic and physiographic" criteria, at bottom of the set.

Bowers (1988, p. 132) pointed out that in his vegetational subdivisions Shreve used geographical titles rather than floristic ones, and that ecologists since have followed Shreve's lead. Bowers also remarked that Shreve's criteria were primarily physiognomic. In this regard he followed such well-known preceding plant geographers as Humboldt (1807) and, more so, Grisebach (1838), Warming (1909), and, as always, Clements (1916a). Even Clements? Yes. Clements' association was the floristic subdivision of a physiognomically-developmentally derived formation. Warming (1909), in what many students interpret as the first "real" Ecology textbook, used growth forms as a basic criteria for physiognomy. (This was reviewed herein: see Range Type, Associations and Consociations, Literature Review.)

Bowers (1988, p. 120) explained that Shreve had been ciritical of the traditional growth- or life forms of Raunkiauer. Instead Shreve used Drude's classification system because it included plant physiological criteria. Herein lies a twist, and one that interwove the careers of Shreve and Clements. Drude was a follower of August Grisebach who in turn was one of Humboldt's main students (Tobey, 1980, ps. 51, 62, 90, 101), and Drude had a tremendous influence on faculty and graduate students at the University of Nebraska including Frederic Clements (Tobey, 1980, ps. 60, 62-69, 71, 83, 87-88, 90, 98-99, 101-102, 204). Though Shreve and Clements disagreed with each other-- usually indirectly or obliquely and with thinly veiled references to the views of the other (ie. flanking actions rather than frontal attacks)-- they had a common intellectual heritage.

Forrest Shreve defied ready made classification. He, having pioneered the continuum concept, was in the individualist school of vegetation, but when he mapped, described, and arranged Sonoran Desert and it's vegetation units Shreve operated at the scale of Clements' associations. Shreve's Sonoran vegetational units or large-scale plant communities (vegetational subdivisions in particular) differed little from those of Clements. Perhaps this can be partially traced to their fundamental philosophy of vegetation that had a common origin in Plant Geography back through Drude. Drude was central to the English or Anglo-American Tradition (Shimwell, 1971, 47-49) which developed into the Clements-Tansley manifestation thereof. Gleason was also a branch-- though a distant one-- from the English Tradition (Shimwell, 1971, ps. 54). Shreve was even more independent than Gleason, and it was studies of vegetation by Shreve-- not Gleason-- that were first to propose the continuum concept for which Gleason got the credit (Bowers, 1988,.ps. 28-29, 54). Again, however, Shreve and Clements had similar pre-desert backgrounds, both having done their initial research in more mesic environments. Also, both had a common perspective of vegetation through the Idealistic Tradition in Plant Geography via Drude (Tobey, 1980, ps. 87-99 passim; Bowers, 1988, 120).

Unlike Lucy Braun's (Braun, 1956) subdivision of the eastern deciduous forest into modified Clementsian associations (Whittaker, 1980, p. 68) and Weaver's (Weaver, 1954; Weaver and Albertson, 1956) treatment of prairie and plains grasslands on basis of Clementsian associations in their purest form, Shreve put maximum distance between his simple arrangement of desert vegetation and the elaborate hierarchial system of Clements that Shreve so detested. The irony was that this difference made little difference. Shreve's treatment of Sonoran Desert vegetation was still based on dominant plant species which was the same basis as Clements' associations which can be dated back through the Third International Botanical Congress and Humboldt, himself, the "inventor" of the association concept (see discussion, Range Type under Literature Review). It is also interesting that the subdivision of the Sonoran Desert on regional and physiographical scale using dominants therein (complete with a version of growth form) was amazingly similar to (though less formal than) the hierarchial vegetation classification system of George Nichols (discussed under George Nichols, Literature Review) which was, of course, similar to that of Clements (ie. yet another "formal treatment of vegetation" in which Shreve took "very little interest").

The other aspect of desert vegetation in which Shreve's ecological views prevailed (to a relative not absolute extent) over those of Clements were on plant succession. Bowers (1988, ps. 103) noted that whereas Clements emphasized the universal occurrence and importance of competition among plants as a factor in succession, Shreve minimized the role of competition in desert vegetation. "In arid regions there has been much less competition. The greatest "struggle" of the plants has not been with one another, but with the environment. Therefore the conditions tending toward the elimination of certain types and the survival and dominance of a relatively uniform one have not been operative. This fact has brought about the diversity of plant types which characterizes many semiarid and arid regions and some extremely arid ones" (Shreve in Shreve and Wiggins, 1964, p. 27). In other words, the element of "competition" that was central to the later-named "relay floristics" of the Weaver and Clements (1938, Chapter VII) textbook scheme was not very important in desert vegetation.

It should be emphasized that it was not only Clements but also some of Shreve's closest colleagues who disagreed with Shreve on this point of competition. McGinnies, who dedicated his book to Forrest Shreve, wrote (dedication notwithstanding): "On the floodplain, competition among the roots was often not as great as on the bajada. The roots of annuals growing on the bajada reach as deep as most perennials in the same habitat, and, since they occur in large numbers, competition with them must be an important causal factor contributing to the sparseness of the perennial vegetation of the bajada. The annuals also come into competition with the shallow rooted perennials through the laterals that are developed on the tap root of the annuals..." (McGinnies, 1981, p. 153).

Shreve wrote that there was "a relative scarcity of plant litter" in the desert. "Consequently this most important of the reactions of plants upon their habitats is reduced to a minimum in the desert. The long existence of a community in a given spot does little if anything to pave the way for the entrance of plants with life requirements different from those of the occupants" (Shreve in Shreve and Wiggins, 1964, ps. 28-29). There was an "almost total lack of reaction by the plant on its habitat" so that presence of a plant does "nothing to make that spot a better habitat for some other plant or some other species" (Shreve, 1942, p. 203; explained further by Bowers, 1988, p. 136). "Them's fightin' words!" Nothing ever written was more anthetical to the Clementsian "development of vegetation" model than to assert that Clements' element of "reaction"-- the factor immediately succeeding "competition" (Weaver and Clements, 1938, Chapter IX and taken from Clements, 1916a, Plant Succession)-- was inoperative in desert vegetation, Clements' "complex organism".

By way of grand conclusion Shreve wrote that "the succession concept would never have been developed in a study of the vegetation of an arid region" (Shrev, 1942, p. 204). Now Shreve had taken his conclusions almost to the point of appearing to having a personal feud with Clements. The wonder is that both men still had a social relationship that was "apparently cordial but superficial" (Bowers, 1988, p. 60). Of course Shreve was sincere and, as shown in later years, correct to a degree. For example, the concept of cyclic succession in the Chihuhuan Desert is today textbook fare (Barbour et al, 1999, ps. 274-275, an example of which was presented herein under Chihuhuan Desert, Shrublands). Emphasis should be laid upon the fact that such vegetation dynamics is interpreted as a form of succession. This again underscores Bowers (1988, p. 88) observation that currently there are "looser definitions" for succession than in the dogmatic days of Shreve and Clements.

It must also be emphasized that Shreve was the ultimate descriptive ecologist. Bowers (1988, p. 86) demonstrated that Shreve appeared to have had an aversion to quantitative methods like quadrants and transects and vegetation mensuration in general. Yet, "he painstakingly measured a variety of physiological factors". "This bias may have arisen in reaction to Clement's advocacy of qunatitative sampling" (Bowers, 1988, p. 86). Here was another irony: it was Clements' invention of the quadrant and his "quantitative method" that decades later is regarded as one of his most lasting contributions to Plant Ecology.

Beyond this, some of Shreve's meticulous measurements of abiotic factors later came into question as to their actual relevance to plant life. Bowers (1988, p. 105) cited correspondence in which Rexford Daubenmire pointed out that Shreve's herculean efforts required to measure simple soil moisture were "completely wasted" because Shreve had not also determined wilting point and field capacity of sampled soil. A defender of Shreve at cross-examination might establish that Daubenmire was a modified Clementsian and product of the Anglo-American Tradition. The other side would point out that this fact was a "rabbit trail".

As to Clements' views, it is a fact that Clements had much less experience than Shreve in the study of desert vegetation. Recall from earlier in this historical note that Clements worked at the Desert Laboratory (in winter) and at the Alpine Laboratory on Pike's Peak (in the summer), and this routine for only eight years. Shreve lived in and studied deserts for over 30 years. Numerous workers have made much of the fact that Clements' ideas of plant succession were formed from his initial work with subhumid to semiarid grasslands and that he brought this experience (with the biases that always come from previous research) when he described "succession" in the desert. The same could be noted of Shreve whose early work was with vegetation in humid Maryland and, later, in the mountain rainforest of Jamaica. Shreve reached the same conclusions about vegetation being a continuum along environmental gradients from his Jamaican studies as from later investigations in the Sonoran Desert. Perhaps both Shreve and Clements over-reacted to the aridity of the desert.

Clements realized that vegetational dynamics in the desert was different from that of mesic and hydric habitats. In Plant Indicators he specified: "No quantitative studies have been made of the actual or potential succession in the desert scrub climax" (Clements, 1920, p. 174). Clements (1920, p. 174) continued in the next paragraph: "The type of succession is peculiar to the desert". Of course, as was explained above, Shreve concluded that there was no such thing as succession in the desert and "climax" was a meaningless concept for desert vegetation (and probably all other vegetation though he stopped short of that, at least "officially").

Clements (1920, p. 174) concluded that Larrea mexicana (= L. tridentata) was the regional dominant and "the Larrea plain is to be regarded as the threefold baseline for topography, climate, and succession, toward which all the others are tending slowly but nevertheless surely". Here in one sentence was the essence of the Clementsian monoclimax in "full-flower". It was the baseline concept of Major John Wesley Powell upon which William Morris Davis based his theory of climatic control of landforms upon which Clements based his theoretical climax formation that would exist when the land surface was eroded down to a peneplain and there was one regional mature vegetation, his climax as a "complex organism". Forrest Shreve rejected this Clementsian paradigm.

In Clements' interpretation the Sonoran Desert was the western desert scrub with the Larrea-Franseria Association (Franseria dumosa= Ambrosia dumosa) while the Chihuhuan Desert was the eastern desert scrub with the Larrea-Flourensia Association. It must be pointed out that Clements (1920, ps. 170-177 ) did not include in his discussion all those units or parts (vegetational subdivisions) that Shreve later mapped and included as Sonoran Desert and that are currently accepted as parts or units (Shreve's vegetational subdivisions) of the Sonoran Desert. Clements (1920) did not describe and discuss those parts existing in Mexico such as Baja California. As such, Clements' coverage extended to only two of Shreve's seven subdivisions (Shreve in Shreve and Wiggins, 1964, ps. 6, 48-49). One of these was the Lower Colorado Valley which Shreve delinated as the Larrea-Franseria Region. This was obviously Clements' Larrea-Franseria Association which was the "Larrea plain" that was the baselevel peneplain (ie. the monoclimax, the regional climax).

The other of Shreve's seven subdivisions that fell within the area covered by Clements (1920) was the Arizona Upland that Shreve named the Cercidium-Opuntia Region. This corresponded closely with Clements' Parkinsonia-Fouquiera community (Parkinsonia microphylla= Cercidium microphylla) between which and the Larrea-Franseria Association there was a Larrea-Parkinsonia ecotone. "The dominants of the foothills and the upper bajadas approach Larrea in requirements... Here the sequence is more difficult to determine because of the irregualr topography and the confusing effect of temperature [Yes, Clements not Shreve wrote this. But Shreve wrote almost exactly the same thing in his version.] ... it may be said the Fouquiera stands nearest Larrea, Cereus [= Carnegia] comes next, and Parkinsonia micarophylla is last. ... Parkinsonia mixes with Larrea in scores of places..." (Clements, 1920, ps. 174-175).

Inescapable conclusion: for all of the personal (though hidden) antimosity and jealousy that seemed to exist between Shreve and Clements and regardless of the different styles of explanation and "oceans-apart" views on formal classification of vegetation, succession, role of competition and reactions, etc. between these two highly respected ecologists, they came up with about the same vegetational units at the scale that corresponds to vegetation cover (= dominance) types.

In his map of potential natural vegetation A.W. Kuchler (1964) used the Shreve units: K-36 (Creosote Bush-Bur Sage, Larrea-Franseria) and K-37 (Palo Verde-Cactus Shrub, Cercidium- Opuntia). The Society for Range Management (1994) used these same units for rangeland cover types: SRM 506 (Creosotebush-Bursage) and SRM 507 (Palo Verde-Cactus). To further illustrate the consistency among these various workers the Chihuhuan Desert was designated as SRM 508 (Creosotebush-Tarbush), K-53 (Trans Pecos Shrub Savanna), and Chihuhuan Desert Scrub by Brown and Lowe with the U. S. Forest Service (later published as Brown, 1994). For the Sonoran Desert (designated Sonoran Desert Scrub) Brown and Lowe (in Brown, 1994) used the seven subdivisions of Shreve when they mapped the greater southwest region of North America.

The Sonoran Desert as interpreted by Shreve (Shreve and Wiggins, 1964, p. 6, 48-49):

1. Microphyllous desert Larrea-Franseria region Lower Colorado Valley

2. Crassicaulescent desert Cercidium-Opuntia region Arizona Upland

3. Arbosuffrutescent desert Olneya-Encelia region Plains of Sonora

4. Arborescent desert Acacia-Prosopis region Foothills of Sorora

5. Sarcocaulescent desert Bursera-Jatropha region Central Gulf Coast

6. Sarcophyllous desert Agave-Franseria region Vizcaino Region

7. Arbocrassicaulescent desert Lysiloma-Machaerocereus region Magdalena Region

Shreve's life-forms were applied to and described those of the physiognomically dominant species. Examples of meanings of life-form terms included: sarcocaulescent= succulent or other forms of thick (pachycaul) trunks (stems), crassicaulescent= crassi (thick) aboveground stems, microphyllous= small leaves, and sarcophyllous= succulent or thick leaves.

Bowers (1988, ps. 130-135) summarized Shreve's seven vegetational subdivisions in which she cited recent ecologists who felt that only three of these subdivisions were actually desert: Lower Colorado Valley, Central Gulf Coast and portions of the Vizcaino Region. The other subdivisions (thereby including the Arizona Upland) were better described as "depauperate thornscrub" in the view of some workers. Even Shreve's treatment of the Sonoran Desert (the subdivisions are among his best known and most lasting contributions) was perhaps colored by his relationship with Clements (Bowers, 1988, ps. 134-135). Shreve's much greater and longer experience with deserts enabled him to provide a more precise and complex meaning of desert than that of Clements based on the latter's limited desert interlude. Bowers (1988, p. 60) wrote that "it is possible that Clements inadvertently compelled Shreve to define his position on competition, vegetation classification, succession, and so forth, particularly in relation to deserts".

This is but another example of the monumental impact of Frederic Clements. Even Bowers' biography of Forrest Shreve is partly a story (partial biography) of Clements. [It should probably be remarked in passing that Bowers (1988) appeared to your current author to have a deeper overall understanding of Clements and the Clementsian paradigm than did Tobey (1981). Bowers (1988) certaily did not review the entire career of Clements but neither did Bowers (1988) make major erroneous conclusions the way Tobey (1981) did. Students were again referred to the Literture Review hererin.]

At any rate, there was another bit of irony (and hypocrisy if truth be told) in Shreve's still-used vegetational subdivisions of the Sonoran Desert. Not only did Shreve designate large-scale units of vegetation that were very similar (identical at some points) to the more superficial units (eg. associations and consociations) of Clements, Shreve also wound up using-- in one of his most familar and important additions to desert vegetation-- a similar form of elaborate jargon and formal titles of plant communities that he critized (always indirectly) Clements for! With Shreve's seven Sonoran Desert subdivisions Clementsian coinages like prisere, subsere, clisere, subclimax, disclimax, preclimax, postclimax, ecesis, aggregation, etc. and Larrea-Franseria association or Parkinsonia-Fouquiera community were joined by the Shrevian coinages of microphyllous desert (Larrea-Franseria region), crassicaulescent desert (Cercidium-Opuntia region), ... and arbocrassicaulescent desert (Lysiloma-Machaerocereus region). Even the term "region" had a parallel in the elaborate Clementsian system in which Shreve took "very liittle interest": "The climax formation is thus a product of reaction operating with ithe limits of the climatic factors of the region concerned". From the perspective of life-forms Clements' climax was the "mature" regional vegetation "... which terminates in the highest life-form possible..." (Clements, 1916a, p. 125).

The tenet of the orthodox Clementsian "dynamic vegetation" (succession-climax) model that Shreve appearently found most objectionable was the "genetic relation" of one climax to another (Shreve in Shreve and Wiggins, 1964, p. 29). An example would be Clements' interpretation that the Cercidium-Opuntia region (large-scale community where these two genera were physiognomic dominants) or Carnegia-Encelia community was a postclimax to the Larrea-Franseria region (= Larrea-Franseria association) as shown in Weaver and Clements (1938, p. 536) or when in Plant Indicators Clements (1920, p. 174) interpreted the Larrea plain as the climax to which all Sonoran Desert vegetation was "tending slowly but nevertheless surely". Such a generalized conceptual view (a unifying theory for land forms, soil formation, and vegetation development) of "dynamic vegetation" at geologic time scale and regional spatial scale did not change the fact that both Clements and Shreve described almost the same plant communities for current physiographically distinct vegetation (eg. that of bajadas vs. valleys or Colorado Valley vs. Arizona Upland).

Aside from differences in interpretation of vegetation based on it's development (Clements's abstract theory or, perhaps more aptly, philosophy), Shreve's view of desert vegetation that distinguished him from strict Clementsians was his individualistic (vs. Clements' organismic) concept of plant communities as a continua along environmental gradients. In a particluarly important passage (quoted above) Shreve wrote that "[e]ach habitat in each subdivision has its own climax" (republished as Shreve in Shreve and Wiggins, 1964, p. 29). This was, of course, Shreve's continuum concept of vegetation (the individualistc view at which he beat Gleason to the punch), but it was (or could be interpreted as) Shreve's equivalent of the polyclimax theory of Tansley or the climax pattern theory of Whittaker.

With that one statement enclosed within a larger sentence (first wrritten in 1942) Shreve stated the range (= habitat) site concept. It was explained in the Literature Review (Range Site) that the range site unit proved acceptance of the Tansley polyclimax and rejection of the Clementsian monoclimax concept, but there was an interesting twist in this. Management of vegetation at the range site scale was probably first stated in the scientific literature by Dyksterhuis (1949), but the basis of this and the quantitative use of certain species to determine range condition had it's origin in the grazing type and plant indicator concepts stated by Clements (1920) in Plant Indicators.

In writing that "[e]ach habitat in each subdivision has it own climax" Shreve (in Shreve and Wiggins, 1964, p. 29) was speaking of Sonoran Desert vegetation within the vegetational subdivision level (ie. at the individual habitat scale-- habitat or range site-- "inside" the larger spatial scale of subdivision). The larger, the subdivision, scale was the only scale at which Clements in Plant Indicators (Clements, 1920, p. 174-175) described Sonoran Desert vegetation, and (as mentioned above) he did this for only two of the seven Sonoran Desert vegetational subdivisions. At this scale (Shreve's vegetational subdivision and Clements' climax) there was not a meaningful difference between the vegetation described and named by either ecologist in his classification system.

There was a certain humor in this "holier than thou" skirmish over sanctity of terms and "real world" fit of concepts. It was rather reminiscent of the event of building the first transcontential railroad in which for some distance both the Union Pacific and Central Pacific laid parallel tracks (rather than joining up) in order to capitalize on land grants and other financial bonanzas. In the case of Shreve and Clements there was a publication bonanza but no act of Congress that forced a Promontory Point meeting of rails. Instead the tracks of Vegetation Science ran asymptomatically parallel. They still do.

In the following presentation, Sonoran Desert vegetation was shown at varying spatial scales ranging from microsite to range site (scale of soil series or association and local physiography) up to scale approaching that of landscape units. These varying scales of the greater Sonoran Desert ecosystem were described in the context of range cover types (and the spatial scale thereof) to stay within the subject of this publication. The photographs were used to represent range dominance types as mapped and/or described by various authorities and to explain various things about Sonoran Desert range ecology. The photographs were not intended to depict representative examples of smaller ecological or vegetational units like range sites or microsites although some of these units of scale were obvious in the photographs. Different spatial scales, aspects, slopes, etc. were easily distinguised in some slides (eg. upper bajada, lower bajada, and valley or north slope and south slope within the overall basin and range physiography). Differences in corresponding vegetation were often dramatic.

If Sonoran Desert vegetation is a continuum along environmental gradients as Shreve-- the undisputed definitive Sonoran authority-- argued, it could be further argued that units of vegetation at smaller scales (eg. range site) are largely meaningless (ie. each is a unique assemblage of species determined primarily by a unique combination of abiotic factors existing along any number of intersecting gradients). At the scale of range cover (= dominance) type there was almost unanimity-- certainly close agreement-- among the published sources.

For vegetation units within Shreve's vegetational subdivisions, the series (Brown, 1994) were used (eg.Creosotebush-White Bursage series and Saltbush series in Lower Colorado River Valley subdivision; Paloverde-Cacti-Mixed Scrub series and Jojoba-Mixed Scrub series in Arizona Upland subdivision). The Brown (1994) series of the Mojave Desert (Mohave Desertscrub) were also used. For some examples, the range sites of Humphrey (1960) were also shown (when it could be determined that these range sites applied).

Given the inherently arbitrary distinction of deserts (especially the border between the Mojave and Sonoran) and the often confusing designation of a "Colorado Desert" (Jaeager, 1957, p. 85) some examples of vegetation were included under both Mojave and Colorado Deserts and Sonoran Desert in this publication. Readers are free to draw their own "property lines".

4. Jojoba or goatnut (Simmondsia chinensis)- This rather unique species has been placed in either of two families (Simmondsiaceae or Buxaceae) by different taxonomists. Judd (1962, p. 93) stated that "California jojoba is one of the most important browse plants in the Southwest". The Forest Service (1940, p. B148) rated jojoba as good to very good winter browse and fair summer browse for all livestock classes. Jojoba is monecious.

Sonoran Desert, Cochise County, Arizona. December.

7. Incienso or white brittlebush in bloom- The farmer "makes hay while the sun shines", but desert plants bloom "if and when it rains". Brittlebush was shown here blooming in the Sonoran Desert in winter (January). The general precipitation pattern of the Sonoran Desert is biseasonal with the major moist periods being winter and summer. (The Sonoran Desert is located between the Chihuhuan Desert which gets most of it's moisture in summer and the Great Basin Desert which gets most of it's precipitation in winter. The Sonoran Desert receives some moisture from the storm tracks and precipitation patterns of each of the neighboring deserts.)

Pinal County, Arizona. January.

8. Inflorescences of white brittlebush or incienso- White and yellow are the two predominant colors of the huge composite family. Brittlebush is readily identified (though not readily by it's flowers), but that is not generally the case for many of the composite forbs. It is common practice among rangemen and foresters when asked the identity of a blooming composite that they do not know to reply with the tongue-in-cheek answer, "Some DYC (or DWC)" (codifying Damned Yellow Composite or Damned White Composite). Viewers see a bright DYC blooming in winter in the Sonoran Desert.

Pinal County, Arizona. January.

11. Creosotebush-white bursage (= burroweed) flats in Colorado Desert- This desert basin was co-dominated by two of the most common scrub species of the Colorado Desert. These were featured in the immediate foreground: creosotebush (left) and bursage or burrow bush (right). This landscape was a good example of the physiography characteristic of the Basin and Range province (a basin in foreground; a mountain range in background). Herbaceous species were largely absent from this browse range. The dried herbage was red bromegrass.

The palatability of creosotebush to larger herbivores like ruminant livestock and wildlife and horses is zero "for all intents and purposes", but white bursage is often a valuable browse plant. Dayton, 1931, p. 154) cited observations that reported white burrow bush to be the most palatable shrub in the region for horses plus being nutritious to cattle and sheep. Stubbendieck et al. (1992) also stated that burrow bush was "preferred by horses and donkeys".

Pinto Basin, Joshua Tree National Monument, Riverside County, California. June. FRES No. 30 (Desert Shrub Ecosystem). K-36 (Creosotebush-Bursage). SRM 506 (Creosotebush-Bursage).

14. Desert wash vegetation- The Colorado Desert has the physiography of the general Basin and Range province of which it is one part. This landscape has pronounced dendritic drainages and such smaller land forms as alluvian fans, benches, arroyos (desert gullies), and stream beds. Many of the latter are ephemeral water courses that are dry much or most of the year. They have often been described as desert washes. Increased quantity of available water, even for brief periods, in these washes supports vegetation that differs from that of adjoining plant communities in species composition, structure, etc. These desert wash communities could be interpreted as riparian vegetation though some might argue that such a perspective is more imagination than interpretation. The general dominant of wash vegetation in the Colorado Desert is the leguminous shrub or small tree commonly known as smoketree (Dalea spinosa= Parosela spinosa). Common associate species include palo verde or green tree (Cerdicium floridum= Parkinsonia floridum), desert ironwood (Olneya tesota), and Fremont cottonwood (Populus fremontii).

Vegetation shown in this slide was a smoketree consociation, but jojoba (Simmondsia californica) as the associate species was also plentiful (by desert standards). Two jojoba plants can be seen here (the dark green, tall shrub in left center and the spreading shrub in far right background). Jojoba as a range browse plant was discussed in two time-honored USDA publications (Dayton, 1931, ps. 94-95; Forest Service, 1940, B148). The current author can personally attest to the value of jojoba as cover for California quail (Callipepla californicus). He "bird-dogged" a nice covey from one of the pictured jojoba plants back to the other. As seen in the photograph herbaceous species were absent from this wash.

Joshua Tree National Monument, Riverside County, California. June. FRES No. 30 (Desert Shrub Ecosystem).Subunit of K-36 (Creosotebush-Bursage). The water course variant or subdivision of SRM 211 (Creosote Bush Scrub).

16. Inflorescences of smoketree- Smoketree is a papilionaceous legume, a member of the Papilionoideae subfamily. The species of this subfamily are often thought of as the "true" legumes because they are the ones with the papilionaceous flowers that are characterized by five petals: the largest and unpaired petal known as the banner or standard, two keels (partially fused), and two wings. Some taxonomists have concluded that this subfamily is the "real" bean or legume family (Leguminosae= Fabaceae) and that the other subfamilies (Mimosoideae) and Caesalpinioideae) should be elevated to the distinct and separate mimosa family (Mimosaceae) and senna or caesalpinia family (Caesalpiniaceae). There is general agreement that it is only members of the papilionaceous group that host nitrogen-fixing bacteria in sufficient populations to add much nitrogen to the soil. Thus papilionaceous legumes are more or less synonymous with nodulated legumes.

Both of these two slides were of the same tree. Joshua Tree National Monument, Riverside County, California. June.The Lower Colorado Valley is the largest and most "desert-like" subdivision of the Sonoran Desert. Creosotebush is the dominant over the vast desert "super-region" of North America from the Chihuhuan through the Great Basin Deserts. Creosotebush is especially singularly prominent over the Lower Colorado portion of the Sonoran Desert.

17. Creosotebush flat (exterior view)- Example of the "Larrea plain" described by Clements (1920, p. 174) and the intermont plains of the Lower Colorado Valley subdivision of Shreve (Shreve in Shreve and Wiggins, 1964, ps. 49-50, 57-58). This is a creosotebush consociation with essentially no white bursage present (author did not find a single plant of bursage). Understorey was dominated by red brome (Bromus rubens) with fluffgrass (Tridens pulchellus= Erioneuron pulchellum) an associate or local do-cominant. There were periodic patches of Arabian or Mediterranea grass (Schismus arabicus). Fluffgrass is a native perennial, but it's stolons and their daughter plants are of such short life as to appear annual-like. Red brome is one of the least productive and -palatable Bromus species (although it will be eaten when young). Arabian grass was described by Hitchcock and Chase (1950) as "an excellent forage grass in winter".

The community shown here was obviously winter range, with the usefulness of that varying from year-to-year depending primarily on winter moisture. In describing creosotebush communities like the one shown here Humphrey (1960, p. 55) remarked that unless they "... may support a crop of winter or summer annuals in season, they could not even be classed as range".

This range community was part of the "Mojave Desert-Sonoran Desert transition", a segment of the Lower Colorado River Valley subdivision that was part of the "tenuous border with the Mojave Desert" (Brown, 1994, p. 190). It could logically be included with the Mojave Desert and, in point of fact, vegetation shown here made a case for the less common, more informal, and less accepted classification of this part of the southwestern desert region as the "Colorado Desert". Brown (1994, p. 157) remarked that the Colorado Desert was "Shreve's Lower Colorado subdivision of the Sonoran Desert". Kuchler (1966) designated this vegetation in lower California as unit #35, Creosote bush (Larrea), in contrast to unit #36, Creosote Bush- Bur Sage (Larrea-Franseria), farther east into Arizona.

Readers attention was directed to the large nest of California harvester ants (Pogonomyrmex claifornicus) in left center foreground. The other mounds were locations of former creosotebush plants (one shown closer in succeeding slide).

Vegetation in this photograph was in an intermont basin of the Sonoran Desert section of the Basin and Range physiographic province, specifically a bolson which is an undrained basin (Fenneman, 1931, ps. 357-377).

Bureau of Land Management, San Bernardino County, California. June. FRES No. 30 (Desert Shrub Ecosystem). K-35 (Creosote bush). SRM 211 (Creosote Bush Scrub). Brown (1994, p. 162-163) included this as the Creosotebush series of Mohave Desertscrub while the scrub community co-dominated by creosotebush and white bursage was designated as the Creosotebush-White Bursage series of the Sonoran Desertscrub. The Kuchler (1966) vegetation map did not delinate deserts so the Mojave and Sonoran were not differentiated.

18. Interior of a creosotebush desert flat- Two views inside the creosotebush scrub range community presented in the immediately preceding slide. These slides showed the regular or uniform dispersion pattern typical of many arid shrublands-- the deserts or desert scrub-- (see Barbour and Major, 1995, ps. 840-841, 875). The understorey was a single herbaceous layer of annual red brome and perennial fluffgrass with periodic patches of the annual Arabian or Mediterranean grass. Both annual grasses were Eurasian natives of anthropogenic introduction. The grasses and a few other species of winter annuals or spring ephemerals had already disintegrated under action of dessicating winds and herbivory (recall harvester ant colony in first slide and note the deer trails in these two photographs: right midground in first slide and central midground of second slide).

The first slide included a mound with remaining woody stems where a creosotebush once grew. This showed the ring growth pattern or form of this sometimes clonal species. Some creosotebush clonal rings have been dated to be several thousand years of age with some rings having several satellite shrubs per clone making distinction of individual "plants" difficult. (Barbour and Major, 1995, p. 837-838). Creosotebush is obviously a xerophyte but it actually has many mesophytic-appearing features (eg. it has C3 metabolism; Barbour et al., 1999, ps. 426, 550 and Barbour and Major, 1995, ps. 881-882). The major adaptation of creosotebush to desert habitats appears to be ability of it's protoplasm to withstand desiccation (Barbour et al., 1999, p. 628).

Bureau of Land Management, San Bernardino County, California. June. FRES N0. 30 (Desert Shrub Ecosystem). K-35 (Creosotebush). SRM 211 (Creosote Bush Scrub). Brown (1994, p. 162-163) designated this as Creosotebush series of Mohave Desertscrub. Distinct from the dominance type with creosotebush and white bursage as co-dominats.

19. Ground-level of creosotebush scrub community- The grass-dominated herb layer (now largely disintegrated) was beneath a creosotebush shrub layer. Except for the sometimes-present cryptogamic layer this vegetation had a simple two-layer sructure. The creosotebush had a clonal ring growth pattern, but this was not as well-defined as in some cases (eg. mound in first slide above preceding caption).

This site was in the Sonoran Desert section of the Basin and Range physiographic province and could be interpreted as either part of the Lower Colorado subdivision of the Sonoran Desert (ie. "Colorado Desert") or as a part of the Mojave Desert. It was best described as the Creosotebush Section of the Mohave Desertscrub (Brown, 1994, ps. 162-163), but it was typical of much of the Sonoran Desert as well, delinations being as mentioned arbitrary.

Bureau of Land Management, San Bernardino County, California. June. FRES No. 30 (Desert Shrub Ecosystem). K-35 (Creosotebush). SRM 211 (Creosote Bush Scrub).

20. Desert saltbush (= allscale)-creosotebush range in Lower Colorado Valley- In what is probably the most comprehensive treatment of vegetation in southwestern North America, Brown and Turner (in Brown, 1994, ps. 194-197) recognized and described a Saltbush series of the Lower Colorado River Valley vegetational lsubdivision of the Sonoran Desert. This was consistent with early (and still outstanding) work by Humphrey (1960, p. 54) who placed both desert saltbush and creosotebush range types as well as "paloverde-triangleleaf bursage sites" (Humphrey's use of "site") and "desert wash" in the "southern desert shrub" (Humphrey, 1960, ps. 52-59)..

Note: It was explained above that recent workers in Brown (1994) added details and subunits to the traditionally recognized deserts (and other biomes in southwestern North America). The subunits were designated as "series". Series was a hierarchial level within a digitized classification syustem for biotic communities (Brown, 1994, ps. 303-315 passim). The series unit (Brown, 1994, p. 306) was taken from several sources including the Daubenmire "climax series" based on "major dominants in climax communities" and also the "series' or "cover types" going back to the Society of American Foresters (1954) as well as the natural vegetation approach of Franklin and Dryness (1973). Obviously series is either the traditional or slightly modified dominance type and therefore corresponds closely to forest and rangeland cover types. Application of the "climax series" (and along with biotic communities [= biomes]) to vegetation of the Sonoran Desert was yet another irony to the Clements-Shreve argument discussed in the introduction to the Sonoran Desert. And again, as was more often the case than not, Shreve lost out. By using climax dominance types in a system built on the "biome approach" (Brown, 1994, p. 8-12) the individualistic concept of communities introduced by Shreve and Gleason was, once again, either ignored or made to "suck hind tit" to the organismic concept of Clements. "The biome as a complex organism" (Clements, 1939, ps. 21-25) and biome, formation, and climax were synonyms "when used in the biotic sense" (Clements, 1939, p. 20) vs. "It is not possible to use the term 'climax' with reference to desert vegetation" (Shreve in Shreve and Wiggins, 1964, p. 29).

In their treatment of the Sonoran Desert, Turner and Brown (in Brown, 1994, ps. 193-200) included five series within the Lower Colorado Valley vegetational subdivision of Forrest Shreve. The only one of these series that received much recognition by those who previously mapped and described Sonoran Desert vegetation was the creosotebush-bursage type. The saltbush series refers to desert saltbush that is also known as allscale, cow lettuce, and cattle spinach (Atriplex polycarpa). Desert saltbush was described as "highly palatable" for livestock and big-game and "excellent cover and nesting habitat for quail" by Sampson and Jespersen (1963, p. 66). Humphrey (1960, p. 54) described desert saltbush range and concluded that desert saltbush was "the most palatable and valuable forage plant in most of the area where it grows" (Humphrey, 1960, p. 114). Given the value of this dominant range plant and it's dominance as a "climax series" the saltbush series should be accorded a separate SRM rangeland cover type designation in future publications (ie. Desert Saltbush-Creosotebush).

In the range community shown in this and the next photograph creosotebush was the associate species overall and co-dominant locally (in patches). Western honey mesquite (Prosopis glandulosa var. torreyana= P. juliflora var. torreyana) was also present (eg. left center).

Relict area except as understorey modified by annual exotics. Maricopa County, Arizona. Early estival aspect, June. FRES No. 30 (Desert Shrub Ecosystem). K-35 (Creosotebush) is closest thing but not the same. Likewise closest rangeland cover type was variant of SRM 211 (Creosote Bush Serub) or perhaps of SRM 506 (Creosotebush-Bursage); neither one was descriptive or precise enough. Desert Saltbush (allscale) series of Lower Colorado River Valley subdivision of Sonoran Desertscrub (Brown, 1994).

21. Allscale-creosotebush form of Sonoran Desert scrub- At local scale of this photograph allscale or desert saltbush and creosotebush shared dominance (the latter was at least a strong associate) on this basin within the Lower Colorado Valley of the Sonoran Desert. This was the climax vegetation and an outstanding desert browse range. It also afforded excellent habitat for the California quail (Callipepla californicus).

The herbaceous understorey was dominated by red brome, but cheatgrass (Bromus tectorum) and Arabian grass were well represented. The one conspicuous forb was London rocket (Sisymbrium irio).These Eurasian annuals were not part of the climax, but-- as is the case for the California annual grassland-- they should be regarded as part of the now-potential vegetation (though naturalized and not natural). It was not clear from the literature to what extent native annual and ephemeral species were components of the virgin vegetation, or even if vegetation of the Sonoran Desert (and other deserts) was studied before Eurasian species that had been introduced from the onset of Spanish exploration were already naturalized. From a Range Management perspective, annual herbaceous species generally provide some feed for range animals and afford some soil protection on intershrub spaces. To the extent that these naturalized Mediterranean plants do not in a substantiative way adversely impact range ecosystem structure and function (eg. do not reduce populations or cover of climax and valuable species like allscale, accelerate soil erosion, or induce range retrogression measured as downward range trend) they are either more-or-less "neutral" or somewhat beneficial. Eurasian annuals did not appear to have the major competition and reaction roles of similar species in the shrublands of the Great Basin Desert or adjoining bunchgrass-shrub steppe.

Turner and Brown (in Brown, 1994, p. 194) concluded that the Desert Saltbush series formerly occupied a total area second only to the Creosotebush-White Bursage series. The desert saltbush "climax series" or cover (= dominance) type was the most widespread community in the Gila Valley of Arizona and Coachella Valley of California before it was converted to field crops. Turner and Brown (in Brown, 1994, Figure 119 caption) summarized: "Few unaltered stands of this community remain: most have become irrigated cropland and the rest have been disturbed by livestock with the resulting establishment of such exotic annual species as Mediterranean grass (Schismus arabicus and S. barbatus), red brome (Bromus rubens), and London rocket (Sisymbrium irio)". Humphrey (1960, p. 54) wrote: "Stands of this shrub [desert saltbush] were originally much more widespread in the Salt River Valley on lands that are now under irrigation". "Trails Plowed Under", Charlie M. Russell.

The degree to which entry of Eurasian annual weeds altered the range vegetation presented here was not known. Neither was it known whether the disturbance-induced alternation of the vegetation was beneficial or detrimental when it was used as a grazing (= range) type.

Maricopa County, Arizona. Early estival aspect, June. FRES No. 30 (Desert Shrub Ecosystem). K-36 (Creosotebush-Bursage). Variant of SRM 201 (Creosote Bush Scrub) or perhaps 506 (Creosotebush-Bursage); neither of these was adequate. There should be an SRM rangeland cover type consistent with the desert saltbush type of Humphrey (1960, p. 54) and the Desert Saltbush series of Brown (1994, ps. 194-197): Desert Saltbush-Creosotebush. Saltbush (Desert) series of Lower Colorado River Valley subdivision of Sonoran Desertscrub (Brown, 1994).

22. Interior of desert saltbush range- Desert saltbush (also allscale, cattle spinach) is an extremely valuable browse plant in the Lower Colorado Valley of the Sonoran Desert (the part often designated the Colorado Desert). Allscale is a climax species of the Sonoran Desert and deserves wise use management to insure it's protection and continued existence on desert grazing lands. Humphrey (1960, p. 114) stated: "Desert saltbush is one of the better browse species in the dsert shrub of southern Arizona. Although, in general, it is the most palatable and valuable forage plant in most of the area where it grows, the arid climate results in a low carrying capacity for these ranges."

Most Atriplex species are dioecious. The two sexes of allscale were easily distinguishable in this slide: female plant at full bloom in left midground; rest are male plants.

Relict area in Maricopa County, Arizona. June, early estival aspect. FRES No. 30 (Desert Shrub Ecosystem). K-35 (Creosotebush), but at mapping scale so large as to include another smaller unit of potential natural vegetation. Variant of SRM 201 (Creosote Bush Scrub) or perhaps of SRM 506 (Creosotebush-Bursage), but these are not really correct. Should be an SRM rangeland cover type of Desert Saltbush-Creosotebush. Recognized as a range type by Humphrey (1960, p. 54). Desert Saltbush series of Lower Colorado River Valley subdivision of Sonoran Desertscrub (Brown, 1994).

23. Female plant of desert saltbush or allscale in full bloom- Allscale is one of the most prized browse plants of the Sonoran and Mojave Deserts. Desert saltbush is a climax shrub and physiogonmic dominant of a desert scrub community of the Lower Colorado Valley subdivision of the Sonoran Desert that is second in size to the more commonly described creosotebush and creosotebush-white bursage cover types.

This attractive lady was doing "her thing" during the worst drought (severe to exceptional) in history of Arizona weather records. The native range species will survive and leave their genetic legacy in progeny if given a chance. This specimen was protected in a relict area. Maricopa County, Arizona. June.

25. Vegetation on upper bajada of Lower Colorado River Valley subdivision of Sonoran Desert- Shreve (in Shreve and Wiggins, 1964, ps. 61-63) explained that ascent in elevation from lower bajada to upper bajada within the Lower Colorado Valley subdivision was accompanied by the addition of foothill paloverde (Cercidium microphyllum) or blue paloverde (C. floridum); appearance of giant saguaro (Carnegia gigantea); and increases of ocotillo or coachwhip (Fouquieria splendens), cane or snake cholla (Opuntia echinocarpa= Cylindropuntia echinocarpa), and California barrel cactus (Ferocactus acanthodes= Echinocactus acanthodes) with a commensurate relative decrease in dominance by creosotebush and white bursage (Franseria dumosa= Ambrosia dumosa). Brown (1994, p. 194) gave an example of an ecotone between the Creosotebush-White Bursage series of the Lower Colorado River Valley subdivision and the Paloverde-Cacti-Mixed Shrub series of the Arizona Upland subdivision.

Those descriptions fit the vegetation shown here except that triangleleaf bursage (Ambrosia deltoides= Franseria deltoides) was an associate of creosotebush in a lower shrub layer that formed a general middle layer of vegetation. Foothill paloverde comprised the upper shrub layer which with creosotebush and bursage formed a general middle layer of vegetation. The lower or herbaceous layer had already disentagrated but it was dominated by the Mediteraanean annual grass, red brome. Saguaro comprised the tree layer.

Pinal County, Arizona. June. FRES No. 30 (Desert Shrub Ecosystem). K-36 (Creosotebush- Bursage). Arborescent variant of SRM 506 (Creosotebush-Bursage).

26. Transition Sonoran Desert scrub- Ecotonal vegetation between Lower Colorado River Valley subdivision and Arizona Uplant subdivision of Sonoran Desert. This general range plant community was described by Shreve (in Shreve and Wiggins, 1965, ps. 63-64) as an upper bajada in the Lower Colorado Valley subdivision. Foothill paloverde formed a line at footslope of the hills while creosotebush dominated the lower slope of this bajada above a typical desert basin. The herbaceous understorey was dominated by the Eurasian annual grass, red brome. There were traces of cheatgrass and also (though even less) of Arabian grass.

The leafless, smaller, and less abundant shrubs were individuals of triangleleaf bursage (ie. an associate to creosotebush). This is the typical drought-deciduous condition, and the vegetation was suffereing through a drought rated as severe to extraordinary. Bursage was more obvious in the second of these two photographs (lower left-hand side).

Pinal County, Arizona. June. FRES No. 30 (Desert Shrub Ecosystem). K-36 (Creosotebush-Bursage). Arborescent variant of 506 (Creosotebush-Bursage).

27. Interior of range vegetation on an upper bajada of Lower Colorado Valley subdivision of Sonoran Desert- "Close-up look" of a transition between Lower Colorado Valley and Arizona Upland subdivisions of Sonoran Desert. This vegetation was prominently zonational as well as transitional. A creosotebush consociation (with triangleleaf bursage as associate) that was featured prominently in the foreground gave way to another consociation-- an exclusive community dominated by a single species-- of foothill paloverde, immediately higher in elevation (farther up the hills). Red brome comprised a nearly "pure" herbaceous understorey at lower levels of this bajada. Any winter or ephemeral spring forbs had long since disentegrated. Rangeland vegetation of the composition described by Shreve (in Shreve and Wiggins, 1964, ps. 63-64) as upper bajada of Lower Colorado Valley of Sonoran Desert.

Note: the geologic term bajada was and still is used quite frequently in descriptions of the Sonoran Desert. Much of this usage appears to trace back to Forrest Shreve, Dr. Sonoran Desert. Allaby (1998) defined bajada: "An extensive gently sloping plain of unconsolidated rock debris resting against the foot of a mountain front in a semi-arid environment. Typically it is made up of a number of coalescing alluvial fans laid down by ephemeral streams as their gradients lessen on leaving the mountain zone. Material is also supplied by the weathering of the mountain front.".

The Arizona Upland is the second largest vegetational subdivision of the Sonoran Desert. It is one of the best developed-- most complex and diverse-- vegetation-wise and is probably the best-known due to it's picturesque species especially cactus like the saguaro. This is largely a function of more annual precipitation, more favorable distribution of precipitation, and generally less xeric soil conditions. Due to location of the Carnegie Desert Botanical Laboratory within the Tucson Mountains (at which Forrest Shreve and other pioneer desert ecologists were headquartered) the Arizona Upland is likely the best studied of the Sonoran Desert subdivisions.

The photographs presented below were of Arizona Upland vegetation in the Tucson Mountains of the Sonoran Desert section of the Basin and Range physiographic province. Photographs were often arranged in sets of "photo-quadrants" to show details of the vegetation in local aggregations or groups of species and different age classes. This arrangement seemed to the author to be consistent with Shreve's individualistic view of vegetation as a continuum along habitat gradients. Alternatively it could be seen as providing "snapshots" of a diverse climax vegetation arranged along range sites and microhabitats (ie. a polyclimax perspective).

Topography ("the lay of the land") was one of the major or predominant abiotic factors determining features of scrub vegetation that was recognized by pioneer desert ecologists. Shreve called this "physiographic control of vegetation": "So great is the importance to desert plants of the texture, depth, and surface character of the soil that in every area of adequate size there are differences in the vegetation which are closely correlated with these conditions. The same relations of vegetation to soil are presented over and over again in hundreds of valleys and intermont plains. There are hills or mountain slopes with rock in place and pockets of soil, there are pediments, upper bajadas with shallow, coarse soil, lower bajadas or plains with deeper soil of uniform but rather coarse texture, and flood plains or playas with deep fine soil. The physiographic features of the basin determine the physical character and distribution of the soils, which in turn determine the vegetation." (Shreve in Shreve and Wiggins, 1964, p. 37).

That quote by Shreve could certainly be interpreted by Gleasonians as reflecting the individualistic view of plant communities, but it could just as readily be viewed as being the Tansian polyclimax concept that was used by Clementsians like Dyksterhuis (1949) to blend Tansley's polyclimax with the basic Clementsian paradigm of grazing types and indicator plants (Clements, 1920) to formulate the range site concept. It was also consistent with the climax pattern concept later proposed by Whittaker (1953) that was more specific and amounted to a refined blend of the Clementsian and Tansian concepts."The same relations of vegetation to soil are presented over and over again in hundreds of valleys and intermont plains" rings "over and over again" of a pattern of range sites arrayed along catenas of repeating patterns of soil series. Either way, photographs were arranged below on basis of physiographic location, in particular as to lower bajada, upper bajada, and hilltop.

28. Various views of lower bajada portions of the Arizona Upland of the Sonoran Desert- As indicated by it's designation, the Arizona Upland is found above the lower parts of the Basin and Range province with a distribution along the higher mountain ranges. There are basins, bajadas, and mountains within the Arizona Upland vegetational subdivision and this topographic variation contributes to amazing diversity in physiognomy, structure, composition, and development of the vegetation in it. Differences in habitat (soils, soil water, temperature, wind currents, steepness, aspect, etc.) due to topographic features are pronounced even between the lower and upper bajada. Hence this has been a common basis for distinction of Arizona Upland vegetation. A good summary of such is Crosswhite and Crosswhite (in Bender, 1982, ps. 217-221).

In this series of four slides (and many of the succeeding photographs) differences in vegetation due to location on bajada and aspect (eg. north slope vs. south slope) as well as those reflecting edaphic features (eg. sandy vs. gravelly soils) were often pronounced-- even striking-- and, as described by Shreve in numerous publications, within relatively short distances (over small spatial scale).Vegetation in these photographs was on a lower bajada above the Santa Cruz River Valley. Soils were generally better developed with more favorable moisture conditions.

Creosotebush was present in each of these "photo-plots" with this regional dominant being conspicuous in the first and fourth photographs. Creosotebush plants were typically smaller than those in the more xeric Lower Colorado Valley subdivision. Saguaro were ever-present, and at younger age classes indicating good regeneration. This was featured prominently in the first two "quadrants" of this set. In the first photograph ocotillo (still retaining it's dry season-deciduous leaves in a severe drought) was growing at far-right foreground. Vegetation presented in the second photograph had more triangleleaf bursage than cresotebush. Most of this bursage had shed it's drought-deciduous leaves, but several plants had retained their leaves. These appeared as short plants with light gray leaves (left-center of photograph). The cactus featured in the second photograph (center foreground and scattered throughout) and in the center of the third photograph was desert pricklypear (Opuntia phaeacantha var. discatha= O. discantha).

The third slide of this foursome showed the diversity of cacti in this lower bajada scrub community. Desert botanists and ecologists like Shreve distinguished between the cylindropuntias and the platyopuntias of the Cactaceae. The former were given the taxonomic rank of genus by some taxonomists (eg. Cylindropuntia fulgida for chain-fruit cholla; Cylindropuntia ramosissima for branched pencil cholla). Cylindropuntia was later put ("lumped") back in Opuntia (eg. Opuntia fulgida; Opuntia ramossissima). Shreve used these two nouns, which at the time of his studies were used to designate "sections" of Opuntia, as "life forms" (eg. in Shreve and Wiggins, 1964, p. 45). In this third photograph the platyopuntia life form was represented by desert pricklypear and the cylindropuntia was represented by chain-fruit cholla (left foreground). The saguaro represented the columnar type of life form according to Shreve's scheme (Shreve in Shreve and Wiggins, 1964, ps. 42-44). The barrel type was not visible in the third photograph, but was presented elsewhere as represented by the genus Ferocactus. All of Shreve's cactus life forms were included in his general life form of stem succulents (Shreve and Wiggins, 1964, p. 44). The smaller, leafless, and drab-appearing shrubs scattered throughout fore- and midground was triangleleaf bursage, an "always there" dominant species.

The range vegetation portrayed in the fourth slide included plenty of creosotebush, but ocotillo and triangleleaf bursage were also present along with foothill paloverde (see midground) and saguaro, the other physiognomic dominants.

Viewers should make special note of the lack of an herbaceous layer, and the absence of herbaceous species even as individual plants. If annual or ephemeral species had been present they had disintegrated and vanished. ("For the wind passeth over it, and it is gone; and the place thereof shall know it no more." Psalms 103:16 KJV.)

Saguaro National Park, Pima County, Arizona. June, early estival aspect (awaiting summer monsoon). FRES No. 30. K-37 (Paloverde-Cactus Shrub). SRM 507 (Palo Verde-Cactus). Paloverde-Cactus-Mixed Scrub series of Arizona Upland subdivision, Sonoran Desertscrub (Brown, 1994).

29a. Diversity of species and life forms in Arizona Upland of Sonoran Desert- Creostoebush, the regional defining dominant, was present in abundance throughout this "quadrant". Shreve (Shreve and Wiggins, 1964, p. 42, 46) described the life form of creosotebush as evergreen (perennial leaves), true (woody, nonsucculent) shrub. Old individuals of saguaro, the columnar type life form of stem succulents, marked the "plot" with picturesque distinction. Saguaro were joined with foothill paloverde (the defining dominant of the Arizona Upland vegetational subdivision), the individual tree of which was featured in far-left foreground and was parasitized by desert mistletoe (this parasitic species was featured below). Paloverde was placed in the life form of trees and shrubs with green stems and small drought-deciduous leaves by Shreve (Shreve and Wiggins, 1964, p, 42-43, 47). Chain-fruit cholla, the cylindropuntia life form of stem succulents, was visible in front of the paloverde; immediately to the right (center foreground) was a fishhook or candy barrel cactus (Ferocactus wislizenii = Echinocactus wislizenii), the barrel type life form of stem succulents. Short, multi-stemmed, leafless shrubs were individuals of triangleleaf bursage whose life form Shreve (Shreve and Wiggins, 1964, p. 45) described as nonsucculent, woody, leafy perennial.No herbaceous species were present.

30. The following set of photographs were taken on bajadas just below hilltops, the crests of the mountain ranges. These "photo-plots" showed species composition, life form, and structure of range vegetation on the upper bajada of the Paloverde-Cactus-Mixed Scrub series of Arizona Upland vegetational subdivision, Sonoran Desert (Brown, 1994). FRES No. 30. K-37 (Paloverde-Cactus Shrub). SRM 507 (Palo Verde Cactus). Shallow Soil Palo-Verde--Triangle Bur-Sage Range Site (Humphrey, 1960, p. 57). All "photo-quadrants" were in Tucson Mountains on Saguaro National Monument, Pima County, Arizona (June).

Sonoran Desert scub on upper bajada in Tucson Mountains- Edaphic conditions become less favorable for plant life as soils of the valleys and lower bajada grade into those of the upper bajada and, finally, rocky mountain tops. Soils are generally less developed and have lower moisture contents in ascending the land from valley to hilltop. This is not an absolute however as soils vary in reponse to many factors. Perhaps more importantly, precipitation may increase with increasing elevation. Likewise, temperatures, especially nightly temperatures, are sometimes warmer at higher elevations than at lower elevations (ie. upper bajada vs. river valleys, basins, and even lower bajada). This is of utmost importance in distribution of frost-intolerant species like saguaro.

In some instances, there is but slight variation in desert vegetation due to compensating arrays of abiotic (even biotic) factors. That phenomenon of "variation within sameness" in range vegetation was shown in this set of photographs from the Tucson Mountains.

30c. Biodiversity in a dry land- The gravelly soil on this upper bajada supported a species-rich range community. (Surface texture of this soil should be compared to that of the lower bajada visible in the preceding set of photographs.). Several of Shreve's life forms were represented by cactus and other species : columnar type by saguaro, platyopuntia by desert pricklypear, shrubs with green-stems and small, drought-deciduous leaves by foothill paloverde, and nonsucclulent, woody, leafy perennial by triangleleaf bursage. There were several age classes of saguaro ranging from dead, bare-wood trunks to only-a-couple-of-feet-tall young ones (note the short, young saguaro in center foreground, and with log of an ancient, fallen "grandparent" adjacent to it). For all the species diversity there were no herbaceous species.

This spot was on an alluvial fan coming down out of the drainage from the hills above.