92 THE SAPROLEGNIACE^ OP THE UNITED STATES, 



ment, were fixed with a saturated aqueous solution of picric acid for twenty-four 

 hours, in the earlier part of the work. Later, this treatment was replaced by expo- 

 sure for fifteen or twenty minutes to a hot saturated aqueous solution of corrosive 

 sublimate (HgCl2). This reagent fixes the cell contents without even the slight dis- 

 tortion caused by picric acid, and is strongly to be recommended for such work, as 

 has been done by Hartog ('89a). The whole specimen was, after fixation, washed 

 and soaked in fifty per cent, alcohol, and then stained for twenty-four hours in 

 Grenacher's or Kleinenberg's haematoxylin. After being washed again and passed 

 through graded alcohols and chloroform into paraffin, in the usual manner, the fly 

 with attached fungi was imbedded in paraffin and cut into sections about 7m in thick- 

 ness by means of the Minot microtome. The sections were then mounted serially in 

 balsam. 



The very numerous nuclei carried into the young oogonium with the protoplasm 

 exhibit the structure of the mycelial nuclei, as has been said (Pig. 30). The num- 

 ber of these nuclei bears no relation to the number of oospheres to be formed, except 

 as both are controlled by the amount of protoplasm in the oogonium. In nine sec- 

 tions, including the whole of a young oogonium, about 60/^ in diameter, of JL. apicu- 

 lata, I have counted 175 nuclear structures. With liberal allowance for the presence 

 of parts of the same nucleus in two sections, it is not probable that the oogonium 

 contained less than 100 nuclei ; yet this species rarely produces more than five 

 oospheres in an oogonium. The nuclei remain passive during the formation of the 

 central vacuole, and finally lie distributed through the parietal layer (Fig. 30). I 

 have never been able to see any evidence of division in oogonial nuclei, and believe, 

 with Hartog ('92) and Dangeard ('90), that it does not occur. After the formation 

 of the parietal layer, the nuclei appear to migrate towards each other and to fuse in 

 pairs (Fig. 31) ; and a little later they are seen to be much less numerous and larger? 

 as well as far less deeply stained by haematoxylin (Fig. 32, 33). Indeed, a careful 

 search with well-managed illumination is necessary for their detection. This is due 

 to the fact that their chromatin masses largely lose their characteristic power and are 

 masked by the granular protoplasm, while the nuclear membrane becomes barely 

 recognizable. The space between the membrane and the chromosome, occupied by 

 the hyaline part of the nucleus, is proportionally larger than in the vegetative nuclei ; 

 and it is probably this fact, combined with the faintness of the other parts, that has 

 led Hartog ('89) to attribute to these fusion-nuclei the vacuolated appearance of the 

 young oogonium. That the two conditions are quite distinct, though occurring 

 simultaneously, as Dangeard ('90) has maintained, may be seen in Fig. 32. The 

 observed reduction in the number of the nuclei is plainly due to nuclear fusions (Fig. 



