Greville^ on Asterolamprce. 
43 
seldom filled up with a minute uniform areolation^ as in the 
older species^ but have their inner margin frequently com- 
posed of a row of elongated and sometimes very large cellules^ 
which, combined with the conspicuous umbilical lines, give 
a peculiar skeleton-like aspect to the valve. It is remark- 
able that,, while this areolation is too variable to afford a 
specific character in some species^ it is very constant in others; 
as, for example, in A. liylandsiana, where single^, large, 
lateral, wedge-shaped cellules, render the narrow portion of 
the rays obsolete ; in ^. Ralfsiana, where the cellulation, if 
such it can be called, is produced by repeated divisions of the 
extremity of the umbilical line, the whole closely resembling 
the delta of a river, as represented in a map ; also in A. punc- 
tata and others. The umbilical lines are invariably simple. 
Among these discs we find, for the first time among Aster o- 
lamprce, the centre not unfrequently occupied with a cluster 
of cellules, which in some species^ as in A. vulgaris and 
affinis, are not of the slightest value, being sometimes very 
numerous, at others wholly absent ; while in a few species, as 
in A. marginal a and Ralfsiana, they are invariably present, 
and highly characteristic. 
In the second of the two sections into which I have divided 
these diatoms, we have an important deviation from the 
normal condition of the genus. The substance of the valve 
becomes thick and opaque (always excepting the rays), and 
there is in every case an umbilical nucleus which may be 
solid, or more or less punctate or cribriform^ or irregularly 
areolate. The umbilical lines are robust ; the segments 
smooth or irregularly punctate, or imperfectly cellulate, con- 
veying the idea that the original cellulation was being gra- 
dually closed up, as if by the deposition of siliceous matter, 
as, for example^ in A. cemula and simulans. If I am right in 
uniting the two valves I have figured under the name of 
A. pulchra, punctation may be either present or absent. In 
all the other species of the section the segments are smooth. 
In various species of both sections, the ray-tube (by which 
term I designate the internal channel which passes down the 
middle of the ray) is more or less visible ; but in A. pulchra 
it is so prominently developed as to be not less conspicuous 
than the umbilical lines^ giving a very rich and complicated 
appearance to the valve. Another character also makes its 
appearance in some of the frustules of the second section, in 
the shape of gland-like puncta or pores, often transmitting a 
brilliant light, the nature and value of which I am at present 
unable to determine. In A. pulchra they are situated on one 
or both sides of the extremity of the rays; in A. stellulata, 
