1888.] MORPHOLOGY OF SUTERNUMERARY PHALANGES. 501 



anfl common blastema ' ; and that the syndesmoses, while intimately 

 related to the sheath, are formed, not as ingrowths of the same, but 

 as differentiations of that mass from which the phalanges are derived, 

 and that they differ from these, initially, only in degree of elongation, 



A mode of origin similar to the foregoing has long since been 

 recorded for the parts of the joints in the Mammalia, including Man. 

 KoUiker, quoting Henke and Reyher (6), to whom we owe the obser- 

 vation, likens ^ the changes undergone, not inappro])riately, to those 

 realized in the differentiation of the vertebral and intervertebral 

 bodies. 



These facts, when viewed in conjunction with those adduced in 

 the foregoing section, show that tbe supernumerary phalanx, the 

 normal phalanges, and the syndesmoses are all on a developmental 

 equality. The last named must then be looked upon as structures 

 which, despite secondary changes, would be liable to take on, more 

 or less completely, the condition of a phalanx. Indications of this 

 are not wanting ; for Leydig, in describing the supernumerary 

 phalanx (his 'Zwischengelenkknorpel ') in Hyla arborea asserts (14. 

 p. 166) : " Er fehlt selbst bei Reptilien nicht, wo icli denselben frliher 

 iibersehen hatte : gegenwaitig kenne ich ihn bei Lacerta {L. muralis) ' 

 und Platydactylus mauritanicus." 



IV. INDIVIDUAL AND CLASSIFICATORY. 

 In the course of our work we have met with certain variations 

 and departures from the predominant types afore described. For 

 sake of clearness, we deferred consideration of these while dealing 

 with questions of general morphology ; and we now describe them 

 in detail, discussing the structures in hand in their bearing upon 

 classification. We deal with the several families in that order 

 most convenient to the circumstances of the case. The list given 

 at the head of each family refers only to specimens examined in 

 detail. The measurements include the tail in tadpoles and in the 

 Urodeles, in the older Anura they denote the length of the body 

 from mouth to vent. 



A. PHANEROGLOSSA. 



a. Hylid^. 



Examined : — 



Hi/la arborea (tadpoles), 40-45 millim. 



H. ccerulea, 30, 73 millim. 



H.freiicineti, 25 millim. 



H. lichenata. 



H. peronii, 42"5 millim. 



Nototrema marsupiatum, 40 millim. 

 ' For other details concerning this generally cf. Strasser (18) and Jordan 



- Entwicklungsgescli. cles Menschen, Zweite Aufl. p. 493. " So wie danu abcr 

 diese Hartegebilde deutlich zu werden beginneu, fangen auch die Zwiselien- 

 glieder an eiuen bestiramten Oharakter anzunehmen in ahnlicher Weise, wie 

 bei der Differenziruiig der knorpeligen Wirbel uud der Lig. inter vcrtebr alia'' 



' We are imable to detect any trace of Buch a structure in this species. 



