538 Dr. H. Woodward—On the Pores in Trilobites. 
the Carboniferous Limestone, holding the same position as in the 
other genera alluded to, being seated in the same furrows, but farther 
back from the anterior margin.” 
Dr. T. Oldham, F.R.S., in the Journ. Geol. Soc. Dublin, 1846 
(vol. ili, part 3, p. 189), writes in his description of Griffithides 
globiceps :—“ In the furrows which separate the cheeks and glabella, 
about half way between the front of the eye and the anterior margin, 
I have observed in all the tolerably-preserved specimens which I 
have seen, a small hole or indentation. These are constant and 
therefore obviously connected with the structure of the creature, 
although I cannot offer an explanation of their use. They are similar 
to those noticed by Portlock in his Ampyx Sarsi” (op. cit. p. 261). 
In 1847 Mr. J. W. Salter communicated a paper to the Geological 
Society, “On the Structure of Trinucleus, with remarks on the 
species”? (Quart. Journ. Geol. Soc., 1847, vol. ili. pp. 261-254), in 
which he observes :—‘‘The peculiar perforated border is the most 
interesting part ofthese animals, and I propose to examine it 
critically. 
“The puncta are almost always arranged in radiating rows, three, 
four, or more holes in each row, and these being at equal distances 
they form concentric lines. In T. granulatus, two of these rows are 
separated by a furrow from the rest; in T. seticornis, three are dis- 
tinct from the remaining two or three, by the front rows being sunk 
in a deep concentric furrow. Other modifications take place: in T. 
jimbriatus, the two front rows are turned downwards; lastly, in T. 
ornatus—for by that name we must call T. Caractaci-—the dots occur 
most frequently in quincunx order, i.e. the radiant rows appear 
zigzag and not direct; this appearance is due to the great obliquity 
of the ray. I wish to call attention to this, because I consider it 
enables us to understand the nature of the enigmatical puncta. 
If we suppose a head furnished with a produced membranous 
margin instead of a perforate one, we shall get at the explanation by 
supposing the membrane to collapse at regular intervals, become 
plicate, then perforate, and lastly, separate into linear processes. 
Now we have in Harpes the flat border, with rows of impressed 
puncta, which have not yet perforated the fringe. In Trinucleus 
jimbriatus, we have a plicate border, the thin interstices of which 
have contracted into pores, which is a step beyond the simple per- 
foration in linear series exhibited by 7. ornatus. Lastly, in Ceraurus 
(Acidaspis, Murch.), we have the structure completed, the linear 
processes being quite separated into spines. This structure is not 
anomalous, for the cellular membrane which forms the inner peri- 
stome of the mass passes through an exactly similar course, becoming 
in different species perforate, in others separated into distinct teeth. 
That these perforate or spinous fringes are not essential, but only 
supplementary parts of the head, may easily be shown, by the fact 
that the width of the head, without the fringe, is exactly that of the 
body, and when the animal is doubled up, the fringe projects freely 
on all sides. We still require to find anomalous specimens in which 
all, or some of the above modifications—plications, perforations, or 
