5.0 - * Total weight In = 3.94041 + 2.99254 1CIX

4.0 - 8 weight ** In - 2.77158 + 2.65095 1nX P l 3.0 - lm l n l 2.0 - 8

1.0 -

0.0 - *

-1.0 -

-2.0 - l -3.ot , , ‘ , , , , , , c

-1.9 -1.7 -1.5 -1.3 -1.1 -0.9 -0.7 -0.5 -0.3 -0.1 0.1

Log, of secondary sten basal diameter (cm)

Fig. 1. Relationship between leaf weight, total weight, and secondary stem basal diameter in deerbnuh. N = 60 sterns, P = 0.97for total weight and rz = 0.83 for leafweight on secondary stems basal diameter. Relationships are significayt at p

Describing the Grass

KELLY WAYNE ALLRED ary producers of rangelands. Of all plant families, the Gramineae, The world’s biggest crop is grass. It is found nearly everywhere, or grass family, is undoubtedly the most important and most from hot deserts to inside the Arctic Circle, and it covers more than essential to man’s existence. a fifth of the land surface of the earth. More people eat grass or In view of this, it seems imperative that we are able to communi- grass products than any other plant food. Range managers are cate about grasses. Effective communication lies in a precise and often grass managers, as grasses are frequently the dominant prim- uniform usage of terms, whereas ambiguity makes communication difficult or impossible. The recognition of this principle prompted Author is assistant professor, Department of Animal and Range Sciences, New Mexico State Univ., Las Croces 88003. the publication of A Glossary of Terms Used in Range Manage- The author wishes to acknowledge the helpful comments of Stephen L. Hatch and ment (Kothmann 1974) by the Society for Range Management. an anonymous reviewer. Several years of teaching and assisting courses in agrostology This is Journal Article 914, New Mexico Agr. Exp. Sta., Las Croccs 88003. Manuscript received April 21, 1981. have brought to light a disconcerting confusion in terminology

672 JOURNAL OF RANGE MANAGEMENT 35(5), September 1982 about the grass inflorescence. Efforts to teach, and to learn about represent a synthesis of my own proposals with ideas gleaned from grasses have been hampered by the lack of uniformity in tetminol- Gould 1968, Hitchcock 1951, Benson 1979, and Smith 1977.): ogy dealing with the grass inflorescence. For example, the Manual 1. The grass inflorescence is delimited at the base by the node of ofthe Grasses ofthe United States (Hitchcock 1951), probably the the uppermost leaf. There are no or (except those most consulted reference on grasses in the United States, USES the of the ) present within an inflorescence. This means that same term, , to describe the different of green little bluestem (Schizachyrium scoparium) and broomsedge blue- sprangletop (Leptochloa dubia), St. Augustine grass (Stenotaph- stem ( virginicus, Fig. 1) produce many individual rum secundatum), semaphore grass (Pleuropogon spp.), silver flowering segments, each subtended by a leaf (the spathe), and each bluestem ( saccharoides), and little bluestem (Schiz- a separate inflorescence clustered with the others. achyrium scoparium). Conversely, for two grasses with almost 2. For descriptive purposes, inflorescence terminology for the identical inflorescences, tumble lovegrass (Eragrostis sessilispica) grasses will be based upon the arrangement of rather than and tumblegrass (Schedonnardus paniculatus), the terms panicle . As has been noted, the spikelet itself is technically an and spike are applied. Although the visual recognition and identifi- inflorescence. cation of a range grass may be independent of whatever terms we 3. The terms panicle, raceme, and spike will continue to be may use to describe that grass, the effective organization, correla- applied, but with the following definitions. tion, storage, and retrieval of a body of knowledge in our minds Spike. (Fig. 2). All inflorescences where the spikelets are sessile depends on uniform description and precise terminology. upon the central axis. Most members of the tribe Triticeae produce this type of inflorescence, as well as Hilaria, Lolium, and Nature and Composition of the Grass Intlorescence Microchloa. Raceme. (Fig. 3). All inflorescences where the spikelets are To best understand the nature of the grass inflorescence, we pediceled, or stalked, upon the central axis. This is relatively rare in , must compare it to the inflorescence of other flowering plants and grasses, but characteristic of Pleuropogon, Brachypodium, some contrast the terminology of each. Danthonia, and some depauperate plants of Bromus. . and As currently used for most flowering plants, an inflorescence is Avena. the flowering portion of a plant, or the arrangement of flowers on a Panicle. (Fig. 4). All inflorescences where the spikelets are not floral axis (Abercrombie et al. 1960, Benson 1979, Gould 1968, borne on the main axis of the inflorescence, but are borne on Smith 1977). The term also originally referred to the sequence of branches. This will include most grasses. Obviously, the variation flowering within a cluster., but this connotation is rarely is great enough that descriptive adjectives will be needed to des- emphasized. Based on the sequence of flowering, inflorescencesare cribe accurately the panicle (see #4 below). The inflorescences of divided into two groups, indeterminate and determinate. Indeter- Andropogon, . , most Eragrostis, and most Poa minate intorescences maintain an active at the apex of will be panicles. the flowering , and may continue to grow as long as condi- 4. lnflorescences will not be compound, that is, an inflorescence tions are favorable. This means the oldest (most mature) flowers type will not be composed of other inflorescence types. Rather, will be found at the base of the floral axis (acropetal development). descriptive adjectives will be used to qualify the variation pattern. In determinate infrorescences the apical meristem differentiates a Thus, the inflorescence of green sprangletop (Leptochloa dubia, flower, and any further elongation of the primary axis is Fig. 5) will best be described as “a panicle of racemose primary halted; the growth of the infiorescence is limited and the oldest branches,” rather than “a panicle of .” The terms raceme flowers are found at the apex of the flowering shoot (basipetal and spike will not be used for part of an inflorescence; this elimi- development). nates a major source of confusion among students. Likewise, blue The terms spike, raceme and panicle, as used for nongmsses, grama (Bouteloua gmcilis Fig. 6) will be “a panicle of spicate apply to indeterminate inflorescences. In spikes and racemes, the primary branches,” and hooded windmillgrass (Chloris cucullata, flowers are sessile and pediceled, respectively, on the primary floral Fig. 7) “a panicle of digitate spicate branches.” Although it may axis, and in panicles the flowers are borne on a secondary appear complicated at first glance, the use of modifiers such as rebranched axis. It should be noted that this inflorescence termi- spicate, racemose, unbranched, digitate, whorled or verticillate, nology is based on flower arrangement. As such, a spikelet of grass and pectinate to describe the variations of the panicle adds clarity is technically an inflorescence by itself: a small spike of essentially and precision to the terminology. sessile flowers enclosed by bracts (paleas, lemmas, glumes). And, 5. Occasionally, a main axis (Hordeum, Schizachyrium, Fig. 8), as in all “true” spikes, the sequence of maturation within the or a branch (Andropogon, Bothriochloa) will bear both sessile and spikelet is from base to apex (acropetal). pediceled spikelets. In the first instance, the inflorescence arbitrar- Because of the similarity of flower and spikelet arrangement, the ily will be termed a spicate raceme. In the second, since the inflores- same terms spike, raceme, and panicle have been applied to the cence is of the panicle type, the description may be “a panicle of grass inflorescence. Grasses, however, generally exhibit a determi- racemose branches.” nate growth pattern of the overall flowering shoot, with the most Pohl (1968) uses the term tame to describe such intermediate mature spikelets toward the apex and least mature toward the base conditions. He defines a rame as (p. 242) “an inflorescence branch (asipetal). The arrangement and maturation of spikelets (notflow- which bears some stalked and some sessile spikelets, as in Andro- ers) are not properly referrable to spike, raceme, or panicle. pogon and Hordeum.“This is clear enough, except for the exam- ples. The Hordeum inflorescence contains no branches (as Proposed Terminology for the Grass Inflorescence specified in the definition), is properly termed a spicate raceme, The grass inflorescence is a highly modified, complex structure, and should be deleted from the definition. We must also under- with a different ontogeny than the nongrass inflorescence. The stand that Andropogon is delimited in the strict sense, excluding same terms have been applied to both because of an over-all Schizachyrium, which was formerly treated as a section of Andro- structural similarity. Rather than propose a new set of terms for pogon,. As with Hordeurn, the inflorescence of Schizachyrium either group, clarity and accuracy may be attained by precise contains no branches and is a spicate raceme. With these emenda- definitions and uniform usage of terms. The terminology presented tions to the definition, the termrame may be defined as an inflores- here is meant to be purely descriptive and does not attempt to cence branch that bears repeating pairs of sessile and pedicellate classify inflorescences on the basis of homology or derivation. spikelets. The term is particularly useful when applied to panicu- Rickett (1944) presents an excellent review of the problems asso- late inflorescences of many ; ie., they are com- ciated with inflorescence classification. For the grass inflorescence, posed of rames. I propose we adopt the following restrictions. (These guidelines

JOURNAL OF RANGE MANAGEMENT 35(5). September 1982 673 -8 -ped

ma

-P -ul t

sb ma

P-

-w

k

“1r db \ PS \ & 88 -P P r -4ll r Ul

0 0

Fii 1-9. Diagrammatic representations of selected grass injlorescences. db = dig’tateprimary branches. i = inforescence. ma= main axis. p =. ped = . ps = pediceled spikelet. r = rame. rb = raaemose primary branch s = qikelet. sb = @ate primary branch. ss = sessile spikelet. ul= uppermost leafdelimiting the inflorescence. (Fig. 1) Ciusteredinflorescences of Andropogon tirginicus. (Fig. 2) Generalizedqikeinjlorescenee. (Fig. 3) Generalized raceme infiorescence. (Fig. 4.) Generalizedpanic& injlorescettce. (Fig. 5) Panicle of green sprangletop. (Fig. 6) Panick of blegrama. (Fig. 7) Pam’& of hooded windmillgrass. (Fig. 8) Spicate racemes of Hordeum and Schizachyrium. (Fig. 9) Panicle of big bluestem.

674 JOURNAL OF RANGE MANAGEMENT 35(5), September 1982 6. The term rachis (rhachis) will refer only to the central axis of It is also well to keep in mind that the variation present in grass an inflorescence, whether spike, raceme, or panicle, but not to the morphology, whether regarding inflorescences, ligules, spikelets, axis of a branch of the inflorescence. This is consistent with its or other features, very often exceeds the limits imposed by our usage when applied to leaves. Therefore, Hordeum will disarticu- efforts at description and categorization. We will certainly encoun- late in the rachis, and Bothriochloo in the joints (or internodes) of ter conditions that do not fit clearly into our categories. These are the rame. relatively infrequent, however, and the main patterns are clear. 7. The inflorescences of several grasses will represent a reduc- Careful use of terminology in describing the grass inflorescence tion from a more developed stage. As has been noted, normally will lead to a clear understanding and a correlation of the different paniculate inflorescences of Bromus and Avena are occasionally types found in grasses. Spikes will be found in most Triticeae, reduced to racemes in depauperate plants. Likewise, it is not panicles with spicate branches in most Chlorideae and many Era- uncommon to find just a single spicate branch in blue grama or in grosteae, and tames in the Andropogoneae. It is hoped that teach- eastern gamagrass (), in which two or three ing efforts will be aided by the discussion presented here, and that are typical. Reductions such as these pose no problems in terminol- communciation about grasses can be clear and useful. ogy, as it is usually evident that such inflorescences represent atypical forms. In blue grama with a single branch, for instance, it Literature Cited is clear that the spikelet-bearing structure is a lateral branch and is Aberuombk, M, C.J. Hkkman, and M.L. Johnaon. 1960. A Dictionary not the central axis or rachis. The inflorescence can be described of Biology. Penguin Books, Harmondworth, England. 251 p. without confusion as ‘a panicle of spicate primary branches, occa- Benson, L. 1979. Plant Classification. Znded. D.C. Heath and Co., Lexing- sionally reduced to a single branch.” ton, Mass. 901 p. Consistent patterns in reduction are also easily described. In the Gould, F.W. 1968. Grass Systematics. McGraw-Hill, New York. 382 p. Andropogoneae tribe a common inflorescence is the spicate Hitchcoek, A.S. 1951. Manual of the grasses of the United States. Rev. by Agnes Chase. U.S. Dep. Agr. Misc. Pub. 200. 1051 p. raceme, with both sessile and pediceled spikelets on the rachis, as in Kothmann, M.M., Cluirnun, Range Term Glossary Committee. 1974. A Schizachyrium, Elyonurus, Trachypogon, and . This Glossary of Terms Used in Range Managment. Sot. Range Manage. has most likely resulted from the consistent reduction of all Denver, Colo. 36 p. branches except one. The remaining branch is generally not a Pohl, W. 1968. How to Know the Grasses. Wm. C. Brown Co., Dubuque, lateral structure as in blue grama, but represents the continuation Iowa. 244 p. of the central axis, and the resulting inflorescence may be termed a Rkkett, H.W. 1944. The classification of inflorescences. Bot. Rev. 10:187- raceme. 231. Smith, J.P., Jr. 1977. Families. Mad River Press, Eureka, Calif. 320 p.