ENDOSKELBTAL SYSTEMS OF LIMTJLUS AND SCORPIO. 363 



The ventral longitudinal muscles of the mesosoma are retained, whilst the metasoma 

 is altogether in a reduced or degenerate condition. 



One special feature resulting from the concentration of the skeletal covering of 

 Limulus in two great tergal plates (the anterior and posterior carapaces) is the deve- 

 lopment of muscles connected with the limbs and other sternal parts of the mesosoma, 

 which do not keep within the limits of the segment to which one end of such muscle 

 may be attached, but take a long course forwards so as to receive their dorsal attach- 

 ment either in the prosomatic carapace or in portions of the meso-metasomatic carapace 

 anterior to (in other cases posterior to) the segment of their insertion. Such vertical- 

 oblique muscles may possibly in some cases be regarded as resulting from dislocations 

 of normal vertical muscles, which normally arise, and are inserted in the tergum and 

 sternum of one and the same segment. But, generally speaking, it seems necessary to 

 regard such muscles as new developments, since it is difficult to imagine the steps by 

 which a muscle (apart from a movement of the hard pieces of tergum or sternum to 

 which it was primitively attached) could acquire new attachments outside the segment 

 to which it properly belongs. 



The great entapophyses {Ent^) of the prosoma of Limulus appear to furnish an 

 important instance of the removal of a skeletal piece from association with the segment 

 to which it primitively belonged, and of its incorporation by fusion with a segment jii 

 front of it. Such a transference is familiar enough in the higher vertebrata in the case 

 of the transference of the centrum of the atlas vertebra to the next following segment, 

 where it appears as the odontoid process of the axis-vertebra. Apparently in the same 

 way the entapophyses of the first mesosomatic segment of Limulus have been detached 

 from the rest of the dorsal sclerite of the first mesosomatic segment and have been incor- 

 porated with the prosomatic carapace, thus taking up a position in front of the great 

 tergal hinge to which they were primitively posterior. The attachment of muscles to 

 these large entapophyses, and the serial relations of those muscles and of the muscles 

 attached to the area of the first segment of the mesosoma and of the succeeding meso- 

 somatic segments, is decidedly in favour of the supposition that such a transference 

 has taken place. The attachment of the external branchials of Limulus (20), and of 

 the two most anterior of the mesosomatic dorso-ventrals (12), gives important evidence 

 in this matter. When we adopt the view that the prosomatic entapophyses are really 

 the entapophyses of the first or genital segment of the mesosoma, and that correspond- 

 ingly the following five pairs of entapophyses belong respectively to the five succeeding 

 segments of the mesosoma, whilst the sixth pair of entapophyses belongs to the meta-' 

 soma, the interpretation of the series of muscles attached to them and to the adjacent 

 parts of the tergum becomes intelligible. This view corresponds with that which I 

 had put forward as the result of a study of the hard parts only, in my article "Limulus 

 an Arachnid" (Q. J. Micr. Sci. 1881). See woodcut, fig. 1, of the tergal surface of 

 Limulus in that article. 



