115 
suppose that fishes are degenerate, having once possessed true penta- 
dactyle limbs. Moreover, as is well known, the theories of MULLER 
(26), Huxtey, PARKER and others cannot now be maintained even on 
purely morphological grounds. The subocular arch cannot be homologous 
with the palato-pterygo-quadrate cartilage of Gnathostomes, because, 
as Huxtey and Parker themselves and others pointed out years ago, 
the maxillo-mandibular nerves pass ventrally to the arch instead of 
dorsally, because, as Auuıs (2) shows, the anterior palatine branch of 
the trigeminus runs dorsal to the so-called palatine instead of ventral 
to it, the so-called palatine being a mere anterior extension of the 
“trabecular” cartilage, and because the subocular arch is so evidently 
a mere secondary outgrowth of the cranium (Arııs, PARKER, ÜOLE), 
having no relationship whatever with true visceral arches (PARKER) 
and appearing in development (it is not present in the Ammocoetes 
larva) at a period long subsequent to that when in true Gnathostomes 
the jaws, hyoid and other visceral arches are all well developed (Max 
SCHULTZE and SHIPLEY, 36). The styloid process of Petromyzon also 
cannot be homologized with the hyoid process of Gnathostomes, be- 
cause, as ALLIS states, the hyomandibular nerve passes behind the 
cartilage instead of in front of it. And finally the researches of 
Nea (27) and later authors have absolutely disproved the alleged 
homology between the so-called “lingual” cartilage (henceforth termed 
the “piston ” cartilage-—Busor, 5—to avoid confusion with the 
cartilage supporting the true tongue of higher Vertebrata) and the 
glosso-hyal cartilage of Gnathostomes, by showing that the musculature 
of the former is innervated by the mandibularis nerve and not by 
hypoglossal fibres. 
As I have already stated, Ayers & Jackson and STOCKARD 
have been foremost in recent years in maintaining that, despite the 
disproval of so many previous suggestions alleging homologies between 
the head skeletons of Marsipobranchs and Gnathostomes, the remote 
ancestors of Marsipobranchs must nevertheless have been Gnathostomes 
in type, and they base their conviction on the fact last mentioned, 
viz. that the musculature of the piston cartilage (or cartilages) is 
innervated by the mandibularis nerve. In other words, these authors 
have revived the antiquated view, apparently first promulgated by 
Gunnerus in 1762, that the piston cartilage (or cartilages) represents 
the much modified and displaced mandible of Gnathostomes, the so- 
called hyoid representing all that is left of the quadrate portion of 
8* 
