136 T. II. BUFFHAM ON NEWLY-DISCOVERED PHENOMENA IN 



tialitj being then in the endochrome contained in the sporangium. 

 The curious and invariable loss of one of the isthmi at the base of 

 the lower female frnstule, noticed already, may be intended to give 

 freer play to the sporangia, which undergo a great deal of fric- 

 tion from the surrounding filaments; indeed, they occasionally show 

 a somewhat battered appearance. 



There is considerable elasticity in these isthmi. I have seen a 

 long filament doubled up while mounting in Deane's gelatine, and 

 then gradually unbend itself, and this after eighteen months' preser- 

 vation in glycerine. The male frustules are, in a very few instances 

 absent, but the scars of their attachments can always be detected 

 with a high power. 



I regret I cannot contribute anything towards a settlement of 

 the question as to whether the sporangial or "zygospore-frustule ' 

 contains a number of gonidia which Dr. Carpenter (" Microscope," 

 6th ed., p. 837) thinks may originate the new generation. Some 

 observations of Prof. Smith seem to point to this, and there is a 

 short note by Mr. G. C. Karop — now our Secretary — in our own 

 Journal (Vol. vi., p. 191), in which the writer describes what may 

 have been immature forms possibly developed from such gonidia. 

 I have seen double sporangia where one of the new frustules had 

 evidently been discharged, but this would prove nothing either way. 

 The length of several mature zygospore-frustules I found to be 

 constant at -0059 inch, and the widest filament I found to be but 

 •0051. If the zygospore-frustule at once increased by ordinary 

 self- division, it seems strange that no filaments were found of a 

 width equal to the length of the initial frustule. 



To sum up, then, the salient features of conjugation in ffiiabdo- 

 nema arcuatum we find : 



1. The male frustules are the smallest in size, have the most 

 definite arrangement of endochrome, and are the most readily de- 

 tached. 



2. The female frustules have slightly longer valves, more nume- 

 rous annuli, and have always a wide band near the middle. 



3. Conjugation is always polyandrous, and is effected by the male 

 frustules attaching themselves indifferently to any part of the annuli 

 of the female frustule. 



4. The result of such conjugation is the production of one spo- 

 rangial or zygospore-frustule if only the basal half of a female frus- 

 tule persist ; but, if both halves persist, each will produce a spo- 

 rangium — the two sporangia being in close opposition. 



