672 ME. B. N. PEACH ON THE FAUNA OP THE [Nov. 1 894, 



margins, the elongation being more or less parallel with the outside 

 edge. In Olenellus reticulatus and 0. gigas this elongated pattern 

 is continued inward beyond the genal spines, but ceases at the 

 pleural angles, the pattern on the raised spaces between these latter 

 points and the sides of the occiput being similar to that found on 

 the pleura of the first free body-ring. If it has been shown that 

 there is good reason to believe that what have once been lateral 

 margins have become posterior, the difference between those trilobites 

 whose facial sutures terminate upon the lateral and posterior margins 

 respectively may be more apparent than real. 



No exact homologues of the anterior spines in Olenelloides have been 

 observed. It may be that the rounded anterior angles of Olenellus 

 intermedins, PI. XXXII. fig. 7, represent the places where such 

 spines have disappeared. As I have already shown, the elongated 

 spines recur at intervals of every third segment on the body of 

 Olenelloides, PI. XXXII. figs. 1-3, and there is a strong presumption 

 in favour of the posterior spines of the head-shield being the repre- 

 sentatives of the pleural spines. The idea suggests itself that the 

 genal and even the anterior spines may represent such elongated 

 pleural spines, in which case, if the intervals between them be the 

 same as in the body, the head-shield may represent at least seven 

 original segments. At first sight, it looks as if this arrangement 

 of more pronounced spines at regular intervals were confined to 

 Olenelloides, but there is a slight recurrence of the phenomenon in 

 Olenellus Lapworthi and 0. reticulatus, for while in them the pleura 

 of the fourth and fifth segments are short and like those of the 

 first and second, those of the sixth segment suddenly expand 

 and bear longer recurved spines. Behind this latter segment the 

 analogy ceases, for the sixth is followed by three or four similar 

 segments. 



So far as dorsal spines are concerned, it is an easy matter to 

 correlate them, notwithstanding the fact that some are mere raised 

 tubercles and others enormously-produced spines like those on 

 Holmia Broggeri, Mesonacis asaphoides, M. vermontana, and M. 

 Mickwitzia 1 . Keeping this in view, it is not difficult to see in the 

 hastate telson of Olenellus the homologue of the small pygidium of 

 Holmia and Mesonacis, each being a single segment. In the former 

 case the dorsal spine has been enormously developed and the rest 

 of the structure dwarfed, while in the latter case the corresponding 

 spine is rudimentary. (See fig. 2, p. 673.) 



The question naturally arises as to what could be the function of 

 the spines. Concerning those that fringe the edge of the head and 

 body, it is highly probable that they served to prevent the bearers 

 from sinking into soft, flocculent sediment, and so being stifled. As 

 the genital organs in annelids and most arthropods open well 

 forward in the body, it may be that the function of the enormously- 

 expanded pleura of the third body-segment in Olenellus was to pro- 

 vide space for enlarged genital glands about this region of the body, 

 and, as suggested in the former paper, the form of the pleura and 

 their spines may even denote sex. 



