WITH NOTES ON OTHER SPECIES. 83 



in their completest development are globular in form (Fig. 19), with dense protoplasm 

 and slightly thickened walls. The connection between the members of a series 

 becomes very slight, and they easily fall apart. They may germinate soon after 

 their formation or after a longer time, but constitute essentially a resting state of 

 the plant, as compared with sporangia. Their germination consists in the production 

 of a germ tube or hypha, from which is formed a sjiorangium with zoospores after a 

 brief growth. This distinction should be emphasized, that whereas the zoosj)ores 

 are formed ^(;^7Am the "resting sporangia," the chlamydospores produce them in a 

 distinct germ tube, although the interior of the chlamydospore is often included in 

 the cavity of the sporangium (Fig. 20). 



Schroeter ('69) has described the formation of "gonidia" in Aclilya, but what he 

 saw appear to have been rather " resting sporangia " than true chlamydospores. 

 Walz ('70) observed and figured the latter in a species of Saprolegnia, and Brefeld 

 describes ('61) similar structures. They occur also in AijJianomyces according to 

 Sorokin ('76), and Zopf finds bodies of this nature to be constant and characteristic 

 in Apodachlya pyrifera. But in the last case they are sometimes lateral in position, 

 are never produced in chains, and appear to be the normal resting form of the plant. 

 Well-developed organs of this sort have appeared in a culture of Achlya Americana. 

 (Fig. 18) ; and in one from a tank for tropical plants under glass they were abundantly 

 produced by plants with the sporangia of Achlya, on which they completely replaced 

 the usual sexual organs (Figs. 19, 20). 



In comparing the organs of non-sexual reproduction in the Saprolegniacece, we 

 observe that they do not differ in any essential feature of their origin and formation. 

 They are, in other words, morphologically similar. But, while the office of the 

 sporangia is the I'apid propagation of the species, the chlamydospores have acquired 

 the resting habit, and the spore character in the development of a germ hypha. The 

 " resting sporangia," in their more specialized forms, constitute an evident link between 

 normal spoi'angia and chlamydospores. 



When considering the generic relationships within this family, and the best 

 arrangement for indicating these relationships, one is forced to the conclusion that, 

 notwithstanding the similarities in structure which have caused Pringsheim ('60) 

 and Hartog ('87) to consider the differences of minor importance, yet the constric- 



