294 THAXTEK. 



endospore is laid down, so that the contents is completely separated 

 from the cavity of the origins. The contents is also more fatty and 

 dense, often composed of distinct elements which may be very regular 

 in size and shape (Fig. 9) and might even be mistaken for endospores. 



The chlamydospores, on the other hand, although' they may closely 

 resemble the zygospores, do not appear, as far as I have seen, to pro- 

 duce a continuous endospore; unless the otherwise anomalous Glaziella 

 in which such an endospore is clearly distinguished (Fig. 91) proves to 

 be an exception. For this reason, in a majority of species, the proto- 

 plasm of the chlamydospore and that of the sporophore are continuous, 

 being connected by a protoplasmic isthmus which may remain un- 

 broken even in mature spores (Fig. 46), or may be finally pinched off 

 in the middle by the gradual thickening of the lateral walls. In a 

 smaller number of instances, the separation between the cavities of 

 the spore and sporophore is accomplished at an early stage through the 

 formation of an independent septum, Figures 52-59 and Figure 85. 

 Spores of the latter type have, for the most part, much thinner walls 

 than those of the former, and were regarded by Bucholtz as perhaps 

 young sporangia. An examination of various species and copious 

 material, however, has convinced me that they are homologous with 

 chlamydospores of the first mentioned type. 



The so-called sporangia which have been above alluded to. Figures 

 60-78, which have been associated only with the genus Endogone, are 

 quite unlike the other types of reproduction; and although I have 

 followed previous writers by including them in this genus, there is no 

 evidence beyond a certain resemblance between the sporangiocarp in 

 the one case and the sporocarp in the other, which would tend to 

 confirm the correctness of this reference. These sporangia are termi- 

 nal vesicles, formed in a solid mass at the extremities of branching 

 sparingly septate filaments which radiate more or less definitely from 

 a cushion-like base. The spores which they contain are variable in 

 size, form and number, thin-walled, with dense or fatty contents, and 

 result from a total cleavage of the sporangial protoplasm. They are 

 so characteristic that it would be quite impossible to mistake them for 

 the spore-like masses above mentioned which may occur in zygospores 

 or sometimes even in chlamydospores. 



Baccarini (1903) was of the opinion that these sporangial types 

 should be removed from the Endogoneae and placed in the Mortierel- 

 leae; the sporangium in both cases being separated from the sporangio- 

 phore by a simple septum. As Bucholtz remarks, this disposition 

 appears to be somewhat premature. It must be confessed, however. 



