Pammel — Anatomical Characters of Seeds of Leguminosae. 119 



known. Tschirch uses it to illustrate mucilaginous endo- 

 sperm in his Angewandte Pflanzenanatomie, but, strangely 

 enough, Wigand (286) considered it to be the inner testa. 



The endosperm is not of the same character throughout. 

 In GleditscJiia triacanthos. Cassia marylandica, Trifolium 

 pratense, Medicago saliva, Desmodium canescens, Lespedeza 

 violacea, and numerous others, where endosperm is strongly 

 developed, it is differentiated into three parts. Sempolowski 

 indicates this in some species studied by him. Harz likewise 

 indicates this differentiation. In Lupinus it consists of a 

 single layer, commonly called the aleurone layer, which, as 

 Guignard (355) has shown, is of wide distribution in plants. 

 It is universal in Cruciferae (417 ) but Strasburger ( 446a, 339 ) 

 incorrectly refers to the seed of Capsella as being exalbu- 

 minous, although he correctly refers to the aleurone layer. 



In most of the North American genera studied it is never 

 absent. The aleurone layer is not always sharply marked 

 from the underlying endosperm, e. g., Astragalus canadensis. 

 In Cassia marylandica and Gleditschia triacanthos it is easily 

 distinguishable from the cells underneath and those above. 

 It presents the same characters which I have found in Cruci- 

 ferae, Rhamnaceae, Berberidaceae, and Sterculiaceae . In 

 view of this fact and its wide distribution in other orders, 

 and its great development in seeds with a large amount of 

 endosperm, as in Gramineae, Ste7'culiaceae, and Berberidaceae, 

 the question naturally arises if this layer has not some func- 

 tion other than the mere storage of reserve proteids. Haber- 

 landt (357) has suggested that the aleurone cells produce 

 diastase during the process of germination and may be classed 

 with the digestion glands of insectivorous plants. The more 

 recent researches of Griiss (350), who has done some excel- 

 lent work with Zea mays, seem to leave no doubt that the 

 aleurone layer is a special secreting organ for the production 

 of diastase. The work of Green (342) partially strengthens 

 the results of Griiss and Haberlandt. Green found that ger- 

 mination was much more rapid when a small amount of 

 endosperm was present. 



Reserve Cellulose and mucilaginous endosperm. — 

 Schleiden and Vogel long ago observed that^the-eiidosperm of 



