866 Transactions of the Society. 



than it has been formerly. Good examples of this diatom may be 

 readily obtained, if picked out, by selection from the sandy sediment 

 of the preparations. It seems to be one of those forms that, from 

 some cause or other, does not remain long in suspension in water, 

 but settles down quickly with the weightier sandy deposit. 

 Trieeratium variahile is now, I think, rarely to be found in guano. 



One of the most decided and remarkable changes that I have 

 observed in the guano of the present, is the complete absence of 

 Aulacodiscus scaher. This diatom was formerly very abundant. 

 Ealfs, in Pritchard's work, 1861, gives it as a new species from 

 Peruvian guano. In every slide of guano of fifteen years past 

 good specimens are very common, but in the preparations of the 

 last eight years, and they have been many, I have not found a 

 single example. Aulacodiscus Comber i is now abundant and seems 

 to supply the place of A. scaher. The processes of this diatom 

 are, upon good authority, said to be " from two to six." I have 

 never seen a specimen with less than three, or more than four, and 

 the frequency of the latter I take to be about one-tenth that of the 

 former. 



A few fine specimens of Aulacodiscus Kittoni may now be had 

 from almost every good preparation. I have found specimens, both 

 single and double, with six, seven, or eight processes. Before 

 mounting this diatom it should be examined when in water, and 

 under an objective of low power. The golden hue of the lines 

 from the umbilicus to the processes is then marvellously beautiful. 



In my last preparation of guano I met with a specimen of 

 A. Kittoni which has fourteen processes. This is a very rare and 

 unusual variety as, I believe, no instance has been previously 

 known in which the processes of this diatom exceed eight in number. 

 It is not a specimen with two valves, each containing seven pro- 

 cesses: such a case I have known, but in that case the seven 

 processes of one valve were only distinctly visible at once, and it 

 was after a considerable alteration of focus that the other alternate 

 seven came into view, and then the others had disappeared. In 

 my rare specimen the fourteen processes and the fourteen double 

 lines of granules from the umbilicus to the processes, appear all at 

 once, in the same plane, and all are equally distinctly defined, even 

 under an eighth-inch objective. I give a figure of this diatom, 

 Fig. 1, Plate XI. 



There is another diatom, ten or twelve examples of which I 

 have found in Peruvian guano, to which I would direct attention. 

 It is an Auliscus, and I believe it to be unmistakably a new species. 

 I found the first specimen of this new form about ten years ago, 

 and then I gave a rough figure of it with a brief notice in ' Science- 

 Gossip.' Since that time I have found it persistently in my pre- 

 parations. Some of the diatoms are single valves, others double. 



