IDENTIFICATION OF THE SEEDS OF SPECIES OF 

 AGROPYRON 



By Robert C. Dahlberg. 

 Seed Analyst, Agricultural Experiment Station of the University of Minnesota ' 



INTRODUCTION 



The identification of the "seeds" - of the species of Agropyron is an 

 important problem to the farmer, the seedsman, and the seed laboratory. 

 Agropyron repens (quack-grass) is recognized as a dangerous weed, and 

 the similarity between the seed of this species and other common but 

 more desirable species of Agropyron gives rise to confusion. Up to the 

 present time no diagnosis has been discovered which appears to be en- 

 tirely satisfactory for use in seed-laboratory practice. This paper deals 

 only with the seeds of the species of Agropyron which are common in the 

 Minnesota seed trade — viz, (i) quack-grass {A. repens (L.) Beauv.) ; (2) 

 western wheat-grass {A. smithii Rydb.); and (3) slender wheat-grass 

 {A. tenenim Vasey). The diagnosis here presented has proved to be not 

 only sound but easily appUed in many hundreds of tests made at the 

 Minnesota Seed Laboratory. 



HISTORICAL REVIEW 



The species of Agropyron are quite easily distinguishable from each 

 other when characteristics of the root systems, leaf characters, spikes, 

 and spikelets are taken into consideration, and these differences are de- 

 scribed in standard works on the taxonomy of the flowering plants (i, 

 2, 3, 4).^ In so far as the seeds are concerned, however, these pub- 

 lished descriptions are not sufficiently detailed for use as a basis for 

 identification of seeds alone. Hillman (5) published detailed descrip- 

 tions of Agropyron spikelets which agree in every way with the observa- 

 tions of the writer, but do not include individual seed characters in suffi- 

 cient detail for an accurate diagnosis of the seeds alone. 



Stevens (8, p. 113) describes in some detail the empty flowering 

 glumes, the size and shape of the seeds, and the rachilla of several species 

 of Agropyron. His descriptions, except as to the rachilla, agree essen- 

 tially with those of the writer. The rachilla is described by Stevens as 

 follows : 



Tlie footstalk (rachilla) in Agropyron repens is entirely smooth, while in Agropyron 

 occidentale * it is rather variable, but commonly with scattered, short, stiff hairs. 



While this may appear to be true when seeds are examined under a 

 low-power lens, yet the rachilla of seeds of A. repens when magnified 



' The writer wishes to acknowIedRC the assistance of Dr. E. M. Freeman, Assistant Dean, and Chief of 

 the Division of Botany and Plant Patholoi^y. and Mr. W. L. Oswald, Chief of the Seed L,aboratory, Min- 

 nesota Experiment Station, in planninE the work and Eiving sURcestions. 



' The word "seeds" is used in this paper in its commercial sense and includes the grain, or caryopsis, 

 inclosed in its persistent glumes, lemma, and palca, with the persistent rachilla segments. 



" Reference is made by number to " Literature cited," p. 281. 



^ Syn. Agropyron smilhii. 



Journal of Agricultural Research, Vol. Ill, No. 3 



Dept. of Agriculture. Washington, D. C. Dec. 15, 1914 



Minn. — 1 



(27s) 



