420 MB. H. M. BERNARD ON THE [Aug. 1 894, 



prostomial position (which has been retained in Limulus) by such a 

 movement of the e3 T es. I further thought that the water-sacs over 

 the eyes of Apus might be evidence of this wandering, these water- 

 sacs being perhaps the integumental fold round the base of the 

 prostomium, which had been drawn back into pockets by the eyes, 

 in which pockets the eyes were consequently situated (fig. 8, p. 421). 

 This sinking-in of the eyes into pockets under the cuticle has been 

 shown by Grobben to be very common among the lower Crustacea. 

 It is found in the Cladocera, Estheridse, Argulus, and in the larval 

 cirripedes, and must, therefore, be considered of very remote origin. 



In comparing the eyes of trilobites with those of Apus, the 

 following points are noteworthy : — 



(a) The eyes, in most trilobites, are found at varying distances 

 from the glabella, on the ' cheeks ' of the head-shield. Olenellus, 

 however, shows that the eyes originally belonged to the glabella, and 

 further, to the first segment. The ocular tubercle in Olenellus is seen 

 branching off from this segment and bending round backwards along 

 the posterior edge of the pleurse of the first segment (see figs. 4 & 5, 

 p. 416). The eyes never cross the great cephalic sutures. 



(b) The fact that, in the trilobites, the eyes wandered laterally 

 off the glabella (which is shown also in the development of Sao 

 liirsuta l ) and took up the most varied positions on the ' cheeks ' of the 

 cephalic shield, seems to show that they had no fixed hereditary locus 

 on the dorsal surface. 



(c) Many of the early trilobites, e. g. Paradoxides, show, in 

 addition to the four more or less clear segmental constrictions between 

 the five segments composing the head, traces of a constriction lying 

 anteriorly, on what is apparently the first segment. If this was a 

 true segmental constriction, then, in these cases, we should have six 

 segments forming the head, which, it must be admitted, is a possible 

 variation. The head-region of Ogygia is apparently, and of Limulus 

 is certainly, composed of six segments; and the secondary fusion of 

 the anterior trunk-segment with the typical number five is what 

 we might expect from the whole process of the formation of the head 

 out of fused segments. But there is another interpretation which 

 requires no more than the normal five head-segments, namely : these 

 anterior infoldings are the openings into pockets into which the 

 eyes have sunk beneath the outer cuticle, pockets homologous with 

 the water-sacs over the eyes of Apus. It is true that in Apus 

 the pore opening into these sacs is unpaired and median. The 

 paired condition of the pores in the trilobites might be due to 

 the wandering apart of the eyes laterally, which has so evidently 

 taken place (cf. Olenellus). These infoldings in the trilobites are 

 nearly always found in a direct line with the eyes, and seem ulti- 

 mately to disappear from the glabella of later trilobites. I would 

 like to suggest, further, that the pores found just in front of the 

 eyes of some trilobites a (pi. xxiv. fig. 30, Barraude), or in the 



1 Korschelt & Heider, ' Vergleichende Entwickelungsgesch.' 1892, p. 512. 

 '-' See Woodward, ' On the Nature of certain Pores observable in the Oephalon 

 or Head-shield of some Trilobites,' App. to Monogr. in Palseont. Soc. vol. xxxviii. 



1884. 



