166 Mr. H. J. Carter on the Conjugations of 



with it, were the largest conjugating elements observed. Frus- 

 tules, apparently of the same species, further down the stream 

 were more than half as large again as the largest size found with 

 the conjugations, which is that above given (fig. 21). 



Thus we observe that all these conjugations are of that class 

 in which two conjugating frustules produce two sporangial ones; 

 moreover that, with the exception of the small conjugating 

 frustules of N. rhomboides, which are terminal, the latter have 

 their conjugating frustules arranged laterally, and not terminally. 

 Lastly, we observe that the sporangial capsules are not retained 

 on the ends of any of the sheaths but those of N. serians. Per- 

 haps this may have arisen from their having been thinner, 

 and therefore more fugaceous ; or, from want of more extended 

 observation on the less developed groups, I might not have 

 observed them. 



Such a capsular covering, however, I suspect to be common 

 to all the spores of the Diatomeae ; and I have also no doubt 

 that, in some of the conjugations already described, the ribbed 

 sheath has been overlooked. It seems to me, now, that I can 

 detect it in my published figures of the conjugations oi Cymhella 

 pediculus ; and I am sure also, by my drawings, that it was pre- 

 sent in some conjugations of Gomphonema which I also found 

 near Budleigh-Salterton in April 1863. But, as I have before 

 stated, these elements are much better recognized among a large 

 number of conjugations which are imbedded in the transparent 

 jelly mentioned than in the naked and occasional occurrence of 

 them under other and less favourable circumstances. 



Formerly I expressed the opinion (Annals, /. c.) that the ob- 

 ject of conjugation in the Diatomese was to restore the normal 

 size of the frustule, which had been reduced by repeated dupli- 

 cative division ; also that one conjugating frustule was always 

 smaller than the other, — citing A. Braun and Meneghini as au- 

 thorities for the former, and assigning my own reasons for the 

 latter. 



With reference to A. Braun^s opinion that the frustules of the 

 Diatomese should decrease in size by duplicative division because 

 it is the case with the cells of Gloeocapsa, &c., I now think that, 

 until this has actually been seen among the Diatomese, the 

 smallness, and therefore variety in 'the size, of their frustules 

 should not be thus explained. 



Again, the statement of Meneghini, that, in Cocconeis, the 

 effect of duplicative division taking place horizontally through 

 this conico-truncate frustule should reduce the size of the frus- 

 tule formed from the smaller end, I also do not now think should 

 be received as an instance of the reduction in size of the Dia- 

 tomean frustule by division, until it has in like manner been 



