ON THE VERTEBRATE SKELETON. 2G7 



broad pleurapophyses are occasionally ossified from two centres in the great 

 land-tortoises of India and the Galapagos isles. The free extremities of the 

 short cervical pleurapophyses of crocodiles and plesiosaurs are expanded and 

 produced forwards and backwards, like axe-blades, whence the name of 

 ' hatchet-bones,' applied to them prior to the recognition of their true homo- 

 logy. 



The pleurapophyses are appended sometimes simply to the ends of par- 

 apophyses ; sometimes to the ends of diapophyses; sometimes by a head and 

 tubercle to both kinds of transverse processes ; sometimes directly to the 

 side of the centrum; and sometimes they are shifted backwards over the in- 

 tervertebral space, and are articulated equally to two centrums (human 

 thorax), and sometimes to two centrums, to a neurapophysis and to a long 

 diapophysis, as in the sacrum of the ostrich (fig. 27, pi). In the atlas of 

 some fishes the pleurapophysis is detached from its centrum, and is suspended, 

 with its haemapophysis, from the antecedent haemal arch (scapulo-coracoid). 

 In some sturgeons the abdominal pleurapophyses are composed of two or 

 more cartilaginous pieces. I have observed some of the expanded pleurapo- 

 physes in the great Testudo elephantopus ossified from two centres, and the 

 resulting divisions continuing distinct but united by suture. The pelvic 

 pleurapophysis is in two pieces, as a general rule (fig. 28, pV attached to 

 D' ) ; and the lower piece is the seat of that most common and simple kind 

 of modification, viz. increase of size with change of form from the cylindrical 

 to a flat bone (as indicated by the dotted line in fig. 27), whereby it comes 

 into connection with the pleurapophyses of other vertebrae besides the proxi- 

 mal piece of its own ; such pleurapophyses having their development stunted 

 so as not to exceed in size the proximal portion of the pelvic pleurapophysis, 

 whose expanded distal portion (02) receives the special name of ' ilium.' This 

 bone retains its rib-like shape however in the chelonians, as in the batrachians: 

 in most species it unites below with two haemapophyses, called, on account 

 of their modifications of form and proportions, 'ischium ' and s pubis.' The 

 pleurapophyses defend the haemal or visceral cavity ; they are the fulcra of 

 the moving powers which expand and contract such cavity in respiration, 

 when its walls admit of those movements ; they frequently support ' diverging 

 appendages,' and give origin to muscles moving such appendages, or acting 

 upon the vertebral column. In some exceptional cases the pleurapophyses 

 become, themselves, locomotive organs, as in serpents and the Draco volans. 



The hcEmapophyses, as osseous elements of a vertebra, are less constant than 

 the pleurapophyses ; although they sometimes exist in segments, e. g. the 

 lumbar vertebrae of certain saurians, and in the case of the ischium, or second 

 pelvic haemapophysis, in which the corresponding pleurapophyses are absent, 

 or short, or anchylosed to the transverse processes. The only true bony 

 haemapophyses in the trunk of fishes appear to be those of the atlas, forming 

 the lower piece of the epicoracoid ; and of the last(?) abdominal vertebra, 

 forming the ischial or pubic inverted arch supporting the appendages called 

 ' ventral fins.' It is at least to the last abdominal vertebra solely that the 

 homologous arch and appendages are connected, by the medium of the 

 pleurapophyses (iliac bones) in the batrachians, and it needs but the removal 

 of the pleurapophysis, or of its second complementary portion (pi 1 in fig. 

 28), to reduce that vertebral segment to the condition which it presents in an 

 abdominal fish. The so liberated inferior (haemapophysial) portion of the 

 pelvic (last abdominal costal) arch is subject, in fishes, to changes of position 

 far more extensive than have been observed in the neurapophyses or pleur- 

 apophyses of the trunk-vertebrae, without however preventing the recognition 

 of the segment to which such shifted haemapophyses actually and essentially 



T 2 



