180 PEOF. W. K. PAEKEE ON THE MOEPHOLOGY 



half-naked between them, and the large basal splint, the parasphenoid (fig. 2, pa.s). 

 This bone reaches more than halfway to the fore end of the long vomers, and nearly to 

 the rim of the foramen magnum ; it is narrow, and for the foremost two fifths deeply 

 split, a small third lobe lying between the two sharp processes. This is a rare condition 

 of this bone, which is only slightly split or gently emarginate as a rule. 



A thin subcutaneous tract of bone binds the suspensorium on the outside. This is the 

 squamosal (sq) ; at present it is very ichthyic, and is mainly composed of the preopercular 

 portion. Several other bones appear afterwards ; we shall find them in tbc skull of the 

 adult specimen. 



The inferior arches (PI. XVII. fig. 4-) are now in their most perfect condition, the 

 three jrill-bearers beincr in full function : thev arc in all essentials like those of 

 Salamandra maculosa (PI. XIV. fig. 3). 



The mandible is lodged in a trough-like articularc (ar) ; a long splenial (sp) runs along 

 the inner side, and the outer face is invested by the dentary (d). 



The hyoid (c.hij) is shaped like a scythe-blade ; it is thickish, especially below. At 

 that end a short hypohyal (//./>//) is segmented off. 



These are the four normal branchial arches, the last of which is functionless. According 

 to rule, the two first only have a ceratobranchial element (c.br 1 ^ 2 ), the rest are merely 

 epibranchials (e.br' &i ). The 1st basal piece (b.br 1 ) receives the 1st and 2nd cerato- 

 branchials by articulation ; the hypohyals arc attached loosely by ligament ; the 2nd 

 basibrancliial (b.br) carries nothing, lies on a lower plane than the 1st, and has its 

 distal end pedate. 



The Skidl o/'Notophthalnius viridescens. Second or adult staye. 



The adult of the specimen dissected by me measured 3j inches from snout to end of tail. 

 I had no other stages to examine but these two, the larva and the adult ; but they are 

 possessed of the true characteristics of an old and a young, or of a low and a high, Uro- 

 dele. This skull is not unlike that of our native large Newt (Triton cristatus), and 

 differs considerably from that of the viviparous Salamander (PL XVI.), being longer 

 and stronger. 



Here, indeed, and in some related types, as in my next instance, Cynops (PI. XVIII. ), 

 we have a very fine piece of skull-building, a bony cranium, forming as safe a structure 

 as could well be provided for any creature whose safety is founded on " the brittle 

 strength of bones." 



It is an almost perfect osteocraniwu, cartilage persisting merely where cartilage is a 

 necessity to this kind of organization. The halves of the occipital arch meet above 

 (PI. XVII. fig. 5) ; but there is a small patch of cartilage on the threshold (fig. 6). 



The small crescentic condyles (oc.c) are but little seen above, as they look principally 

 downwards and backwards ; they are wide apart, having a round notch between them for 

 an intercalary vertebra. 



The ear-masses are large and very strongly ossified ; the large posterior canal (fig. 5, 

 p.sc) runs along nearly parallel with the occipital rim to meet the anterior arch (a.sc). 

 The two arches join at a right angle, and the foremost has a crescentic ridge traversing 



