42 
Greville^ on the genus Auliscus. 
to me that the time has not yet arrived when the introduction 
of a doubtful species is to be regarded as so serious an intru- 
sion. Besides^ as I have ah-eady observed, it requires almost 
as much caution to call a new form a questionable variety of 
some known species, as to put it down at once a questionable 
species. A series of doubtful varieties is exceedingly incon- 
venient by weakening the original definition, especially when 
we are only groping our way to what really constitutes specific 
difference, and consequently a reliable diagnosis in this very 
peculiar tribe. 
Auliscus, Ehr., Bail. 
Frustules cylindrical or discoid ; valve circular or oval, un- 
dulated, with two (three, or four ?) opposite circular, flattened, 
submarginal processes, and four groups of lines radiating 
from the centre; two of them converging towards the pro- 
cesses, and two expanding towards the margin. 
It is remarkable how much uncertainty appears to exist 
with reference to the species of this small genus. This has 
been owing partly to our ignorance in not knowing what part 
of the structure furnished the most trustworthy characters, 
and partly to the very small number of examples in collec- 
tions. So little are the American species of my late lamented 
correspondent. Professor Bailey, understood, viz., A. radiatus, 
pruinosus, punctatus/ Rnd ccslatus, that Dr. Lewis remarks — 
^' They vary much in their markings, and occasionally ap- 
proach so near each other in general character as to make it 
very doubtful whether they ought to be kept apart.^^ This 
want of confidence in Professor Bailey's species has induced 
me to reproduce figures of the whole of them, as those which 
accompany his original descriptions are extremely vague and 
deficient in detail. It is true, however, that some of the Aulisci 
do sometimes resemble each other so closely as to render the 
task of discrimination exceedingly difficult. Until recently 
the number of processes was regarded generically as two. 
A. pruinosus, indeed, not unfrequently occurs with three. 
Within the last two months, however, not only has another 
species been discovered with three, but two species with four 
processes ; one of these will be found described as A. John- 
sonianus ; the other is not in a sufficiently perfect state 
for description, although several valves of it have been seen, 
. all of them showing tolerably distinctly the alternate pro- 
cesses of the subjacent valve; so that the disc is ornamented 
