160 Trans, Acad. Sci. of St. Louis. 



is possible to make out cells with a central globular body. 

 I do not hesitate in calling the secretion reservoir lysigenetic. 

 Tests were made with alcannin for resin, but with negative 

 results. The same reagent indicated an abundance of fat 

 in the other cells of the cotvledons. Check tests were 

 made for other resins, so that there seems to be good evidence 

 that the secretion is not resinous, nor does it contain fatty 

 acids since potash does not destroy the contents of the res- 

 ervoir, but with long-continued action they turn yellowish ; 

 with this reagent the reservoir colors a somewhat deeper 

 brown. The reservoir failed to give tannin reaction with 

 ferric-chloride. Sulphuric acid and iodine color the contents 

 brown, the wall blue. Chlor-iodide of zinc acts in the same 

 way. Delafield's haematoxylin colors the walls violet, but no 

 action takes place on the contents of the reservoir. Fuchsine 

 colors the contents of the reservoir bright red, much deeper 

 than the contents of the parenchyma. 



Whether the secretion receptacle may not contain poisonous 

 products, such as have been found in young branches and 

 flowers of Tephrosia toxicaria, and other species (Radlkofer 

 and Rosenthal), I have been unable to determine. An ex- 

 amination of herbarium material did not reveal similar struc- 

 tures in parts of the plants examined (petiole). The 

 reservoirs of the peduncle resemble those figured for Copai- 

 fera Langsdorfii by Tschirch in his Angewandte Pflanzen- 

 anatomie. The peduncle of Tephrosia virginiana has similar 

 secretion reservoirs. Apios tuberosa^ according to Gray's 

 Manual, contains latex. There is no evidence that the sub- 

 stance in Tephrosia is latex. The walls of the surrounding 

 cells color blue with chlor-iodide of zinc. — (St. Louis, 

 Eggert, Mo. Bot. Gard.) 



Tephrosia hispidula Pers. ( Cracca hispidula, Michx., K.) 



PL XII. f. 2a. 



Testa and endosperm are somewhat more strongly developed 

 than in the last species, 154 /a in thickness. Malpighian cells 

 50 /i long; light line a sharp zone below the cuticularized 

 layer; the cavity irregular in the lower part of the cell. 

 Osteosclerids are longer than in T. virginiana; the inter- 

 cellular spaces are very long. An abundance of pigment is 



