1908] DORETY—CERATOZAMIA 207 



found to have a single cylinder, but peduncular strands occurred in 

 the pith, an observation which was corroborated later by Worsdell 

 (16). In the work previously mentioned (6) Matte made a careful 

 study of the ovulate cone and the mature leaf, with a few observations 

 on younger leaves. He found terminal teeth on some of the pinnae 

 of the younger plants, and at the bases of the petioles stipules with a 

 bundle in each. He says that the meriphyte of the leaves has a 

 modified H-shape in cross-section, with an anterior system, and shows 

 that in this as well as in other cycad petioles the so-called fusions of 

 the traces are often only approximations (accolements) . 



MATERIAL AND METHODS 



The seedlings of Ceratozamia and Dioon spinulosum were grown 

 from seeds provided by Professor Charles J. Chamberlain; the 

 Microcycas seedlings were given to me by Professor Otis W. Cald- 

 well, and later some were grown from seeds secured by Professor 

 Chamberlain. The Zamia seedlings were furnished by Dr. W. J. G. 

 Land. 



Having in mind the danger of drawing conclusions from a few 

 specimens, I have used freely the wealth of material at my disposal. 

 The chief part of the investigation was made from material fixed 

 in picro-acetic-alcohol, stained with safranin and anilin blue, and cut 

 in serial sections. Figs. 21, 22, and 23 are diagrams of wax models 

 constructed from serial sections. 



GENERAL DESCRIPTION OF EMBRYO AND SEEDLING 



The seeds of Ceratozamia are about 2.5 cm long and i.5 cm thick, 

 that is, about the same size as those of Dioon edule, and intermediate 

 between Dioon spinulosum on the one hand and Zamia on the other. 

 Like most cycad seeds, they are flattened on two or three sides by 

 the pressure of growth within the cone, and it is not unimportant 

 to remember that during the whole period of embryonal development 

 — the longest known for a cycad — the seed lies upon whatever side 

 it happens to fall. The single cotyledon begins to be differentiated 

 in November, and early in December its appearance is like that 

 represented in fig. 1. Later it begins to surround the axis (fig. 2), 

 and finally the two edges meet (fig. 3). In these early stages, the 

 coleorhiza is proportionally long; but later elongation of the coty- 



