TYPE IN WHICH THE ADULT PYGIDIUM DEVELOPS LATE. 337 



the surface characteristic of Sao develops. The changes in the 

 position of the optic rudiments (a) in Sao is of great interest in con- 

 nection with the position of the lateral eyes of Limulus. The optic 

 rudiments originally lie quite near the anterior margin of the cephalic 

 shield on each side of the glabella, and at this stage the transverse 

 diameter of the eye is its greatest diameter. Their position in these 

 early stages somewhat recalls the position retained throughout life by 

 the eyes of Cromus inter costatus. Only in the later ontogenetic 

 stages of Sao do the eyes shift laterally and posteriorly away from 

 the glabella, in such a way that their greatest diameter runs parallel 

 with the longitudinal axis of the body. 



The development of Dalmanites socialis follows a precisely similar course. 

 That of Pty 'chop aria Linnarssoni also, which was made known by MATTHEW 

 (No. 4), does not differ essentially from that of Sao. A remarkable feature of 

 the first stages of Ptychoparia is the approximation of the two dorsal furrows, 

 causing the narrowing of the glabella which is not yet sharply defined. In 

 the glabella itself, it is evident that the posterior segments, separated by trans- 

 verse furrows, are at first shorter and more crowded than the long anterior 

 segments, a contrast which disappears as the cephalic region develops further. 

 Here also the position of the optic rudiments changes, in the way de&cribed above 

 in connection with Sao and Dalmanites. 



Special interest attaches to the statements of FORD (N"os. 2 and 3) concerning 

 the ontogenetic stages of the American form Olenellus asaphoides. The youngest 

 stages are here, as in Sao, disc-shaped (Fig. 151 A). The rudiment of the 

 glabella, consisting of five consecutive segments, can be recognised, and behind 

 this a small, still unsegmented region (p), in which lie the rudiments of the 

 whole thorax and pygidium. In the next stage (), this region shows the first 

 traces of segmentation. On each side of the glabella lie the two S-shaped 

 swellings (c, d), which are continued posteriorly into spines (a, b) that project 

 beyond the margin of the body. The outer swelling (c) represents the rudiment 

 of the eye, while the inner (d) takes part in the formation of the "fixed cheek." 

 Of the two pairs of spines which run backwards, the outer (a) probably persists 

 as the "cheek spines." The inner spines (b) are still recognisable in later 

 stages, but then disappear, and in the adult (E] are represented merely by a 

 ridge which runs diagonally from the eyes to the posterior margin of the cephalic 

 shield. During development, a considerable portion of the posterior margin 

 of the cephalic shield becomes intercalated between the two pairs of spines. 

 The inner spine is of interest on account of its position. We are perhaps 

 justified, as will be shown when we come to describe the development of the 

 cephalic shield in Limulus (p. 352), in distinguishing three regions in the 

 cephalic shield of the Trilobites (as also in that of Limulus}, the boundaries 

 of these regions being indicated by the facial suture. We should then have 

 to regard only the "fixed cheeks" as the pleurae of the posterior segments of 

 the glabella, while the "movable" or "free" cheeks, together with the eyes, 

 belonged originally to the most anterior cephalic segment, these latter having 

 shifted by lateral and backward growth round the posterior cephalic segments 

 until they assumed the position in which we now see them. The position of 

 the eyes of Limulus in a posterior, so-called thoracic segment would thus be 



z 



