152 A. COTTAM ON A NEW AULACODISCUS, 



adds — » Sometimes they appear within a faint circle." This des- 

 cription again is somewhat inaccurate, for the hood is not circular, 

 but like the thick end of a wedge with the corners rounded off, and 

 the point of the wedge, if the sides were produced, would be the 

 centre of the valve or very near it. Mr. Hardman seems to think 

 that the hoods may have been injured, some portion of them 

 destroyed by the strong boiling in acids which was requisite to 

 clean the Algoa Bay guano. And he informs me that he has 

 found two specimens from the Cambridge, Barbadoes deposit, 

 mounted some years ago, which are identical with the Algoa Bay 

 specimens; "but the processes are rather more developed, yet 

 much less so than the West Coast of Africa one." 



The new African diatom appears, therefore, to agree with A. 

 Kittoni in the arrangement of its granulation, although the granules 

 - are smaller. It differs from it in generally having a small um- 

 bilicus, and in its processes, which instead of being mammiform, 

 have distinct circular hoods. It appears to me to differ from A. 

 Johnsonii in the arrangement and size of its granules (although 

 size is not of much value as a specific distinction), but especially 

 in the fact that A. Johnsonii has no raised portions under the pro- 

 cesses, and has granules of very different sizes on the same valve. 

 They agree in having an umbilicus, although its presence in the 

 new form is not invariable ; and in these too the form of the pro- 

 cesses is more alike, although more highly developed in the West 

 African form. 



It appears to me that the West African diatom differs in some 

 respects so distinctly from both the species with which it has been 

 connected, that it may be considered a new species, and I propose 

 to call it Aulacodiscus Afncanus, by which name Mr. Martin, I 

 believe, wishes it to be known. I should describe it as follows : — 



Aulacodiscus Africanus. Disc hyaline, with 8, 4, or 5 sub- 

 marginal processes, each covered with a highly developed circular 

 hood. Valve inflated under each process, in the centre usually a 

 small umbilicus surrounded by a rosette of oblong cellules. Granu- 

 lation minute, crowded, radiant, and equal in size all over the valve. 

 Diameter from -0046" to -0021". Becent, Banana Creek, on the 

 West Coast of Africa. 



Figs. 3 and 5 on the accompanying plate have been drawn to 

 show the curious resemblance in that position of two forms that in 

 direct side view are very different ; fig. 3 being the same form as 



