GOO CRUSTACEA side, we shall find there are sufficiently good grounds for placing them in the Edri ophthalmia together. 1 TriloMta (Fossil, extinct). 1 . Eyes sessile, compound. 2. No ocelli visible. 3. (Appendages partly oral, partly ambulatory, arranged in pairs). 4. Thoracic segments variable in number, from 6 even to 26, free and movable, animal sometimes rolling in a ball. 5. Abdominal somites coalesced, forming a broad caudal shield (bearing the branchiae be neath?). 6. Lip-plate, well-developed. Isopoda (Fossil and living). 1. Eyes sessile, compound. 2. No ocelli visible. 3. Appendages "partly oral, part ly ambulatory, arranged in pairs. 4. Thoracic segments usually seven, free and movable, ani mal sometimes rolling iu a ball. 5. Abdominal somites coalesced, forming a broad caudal shield, bearing the bran- chise beneath. 6. Lip-plate, small. Perhaps no investigator of fossil forms has devoted so much careful research to any group as M. Barrande has ex pended upon the extinct Trilobita. Writing recently upon the divisions of their body he arranges them in four groups, according to the number of their free movable thoracic segments. The 1st, of 2 genera, has from 1 to 4 free thoracic segments. ,, 2nd,, 24 ,, 5 to 9 3d 32 10 to 13 4th ,, 16 14 to 26 ,, We thus perceive that the number of those forms of Trilobites which have a great excess of free segments is not large when we consider the group as a whole. In the higher and more specialized forms of Isopoda of the present day, we do not find the number of segments FIG. 74. Trinudeui ornaius, Sternb. sp. (copied from Barrande s Systhne Silurien du Centre de la Boheme, Prague, 1852, 4to, plate 30). Specimens ar ranged in series according to their supposed age. (All the stages figured by Barrande are not given here.) 1. Young individual, destitute of thoracic segments, composed of head-shield and pygidium only. 2. Another of the same stage, in which the genal or cheek spines are developed. 3. Individual with one thoracic segment developed, but without the.genul spines. 4. Another of the same stage, with the genal spines. 6. Individual with two thoracic segments, and with the genal spines present. >>. Individual with three thoracic segments, and possessing the genal spines. 7. Individual with five thoracic segments, but without genal spines. absolutely adhered to without any variations ; on the con trary, we constantly meet with individuals in which more or fewer segments are welded together, so as to conceal the normal number of seven thoracic somites between the head and the abdomen. Such being the case, we cannot be surprised to find considerable variation in a group like the Trilobita, which, if they really are the remote ancestors of the recent Isopoda, must, according to the views suggested above, be the prototypes of the larvae rather than of the adult stage of the living Isopoda. 2 In his researches among the Trilobites of Bohemia M. Barrande has discovered forms which, there is every reason to believe, exhibit (as he has so admirably shown in his 1 H. Woodward, Report on Structure and Classification of Fossil Crustacea, Brit. Assoc. Edinburgh, 1871. The larvae of Bopyrus, Cryptothiria, and Asellus, and the adult JS jidce, Idoteidce, SphceromidcK, and Oniscidce offer many points of analogy with the extinct Trilobita (see the History of the British Sessile-eyed Crustacea by C. Speuce Bate, F.R.S. and J. 0. Westwood, M.A., in 2 vols. 1863-68, 8vo.) great work) the gradual development of the Tiilubite from the earliest form on quitting the egg to the adult. We give seven of the earliest stages of Trinucleus ornatus and seven of /Sao hirsuta, copied from M. Barrande s monograph. 10 FIG. 75. Sao hirsuta, Barrande (copied from pi. 7 of Barrande s work above cited). Barrande figures twenty stages of this trilobite, of which seven are reproduced here. 1. First stage. A young individual in which the limit of the head-shield is not indicated as separating it from the pygidium. 2. Second stage. Young individual with the head-shield separated, and having indications of three soldered segments to the pygidium. 3. Third stage, in which the genal angles of the head and the spiny border of the pygidium are well seen, and four or five soldered segments indicated. 4. Fourth stage, in which two free thoracic segments are developed behind the head, and two or three soldered segments represent the pygidium. 5. Fifth stage, in which the thorax is longer than the head, and is composed of three movable segments and three soldered segments in the pygidium. 6. Sixth stage, in which four free segments succeed the head, and three or four soldered segments form the pygidium. 7. Tenth stage, in which eight free segments succeed the head, and three soldered segments form the pygidium. In the twentieth stage figured by Barrande the adult has seventeen free thoracico-abdominal segments and two soldered ones (the pygidium). One most striking feature in the Trilobita is the remark- abb development of their compound eyes (fig. 11), a subject ably discussed and illustrated long ago by Dr Buckland in his Bridgewater Treatise (1836). Perhaps the eye of the Trilobite may be best compared with that of Limulus, but there are forms like JEglina in which the eyes are enormously developed, occupying nearly the entire head- shield with their facetted surfaces. We have an analogous development of the organs of vision amongst some of the pelagic Amphipoda, the ffyperiidce, and in a very singular form brought home by the " Challenger," the Thaumops pellucida (Phil. Trans. 1873). The " facial suture " in the head-shield of the Trilo bita, which separates the lateral genal portion from the glabella, was for a long time considered as peculiar to Trilobites and Limuli, but C. Spence Bate has ably shown that it homologizes with the suture which traverses the inferior surface in the carapace of the Brachyurous Decapod and the cervical suture in the Macrourau type (Reports Brit. Assoc., Bristol, 1875, p. 46.) The Trilobita are the chief representatives of the Crust acean class in Cambrian times. 3 More than 500 species have been described ; out of these 350, representing 42 genera, have been recorded from the Lower Paleozoic rocks of Bohemia alone by Barrande. About 51 genera and 304 species are British in Cambrian 4 and Silurian rocks ; ten are Devonian, and four Carboniferous. A gigantic Paradoxides, nearly two feet in length, occurs in the Middle Cambrian, and large forms of Asaphus, Homalonolus, Lichas, &c., are met in the Bala group. Phacops, Sphcerexochus, Encrinurus, Calymene, I dcenus, and Acidaspis are among the Upper Silurian forms, some, like Acidaspis, being extravagantly orna mented with spines and tubercles. The Devonian has 8 Agnostus, the earliest genus met with, reminds one of the larval forms of Sao and Trinucleus. 4 The large accession in late years to the fauna of these Cambrian
rocks has resulted from the labours of Mr Henry Hicks, F.G.3.