Homy Teeth of the Marsipobranchii. 385 



If then the nuclear layer be taken as homologous with the 

 stratified epithelial layer, we find that the calcified scales are 

 developed in the deeper layers just as are the horny plates. In 

 the former case the cells immediately above the scale give rise to 

 the scale-pocket, while those immediately below, to the fibrous 

 basis of the scale. Both the scale and the horny plate continue 

 to grow at the periphery, i.e., where the subjacent and superficial 

 layers of cells become continuous, the horn-groove of Beard. This 

 gives precisely the picture of the development of the more in- 

 wardly placed teeth of Petromyzon " without the intervention of 

 a tooth-sac." Towards the edges of the mouth the skin becomes 

 corrugated, involving the cellular layer and giving rise to the 

 appearance of a tooth-sac, in the interior of which would be 

 included the deeper or more fibrous layer of the dermis forming 

 the mesoblastic papilla of reticulate connective tissue. Such a 

 buckling up of the cellular layers would give rise to the conical 

 horny tooth, the centrum of the scale corresponding to the apex 

 of the tooth. Just as the centrum is the first part of the scale 

 to be formed and continues to grow centrifugally, so Warren [3] 

 states that the process of cornification proceeds from the apex 

 downwards. 



If the above suggestion be correct then the horny teeth must 

 be regarded as purely dermal structures and not epidermal as 

 Beard thought. The scales of the Teleostei I believe to be 

 dermal in origin, and therefore the horny plates would be homo- 

 logous with the calcified portion of the scale and, if the homology 

 may be pushed, possibly also to the dentinal layer of true teeth. 



Upon this assumption as to the primitive origin of the tooth - 

 sac it follows that the downgrowth of the oral epithelium in 

 those cases in which it persists, for example in mammals, to form 

 the enamel organ is a secondary feature. This is a point of some 

 practical importance. The difficulty in deciding whether to rely 

 on the shape of the dental papilla or on the contour of the enamel 

 organ, where, as is so frequently the case the two do not coincide, 

 has presented itself to everyone who has worked at the evolution 

 of the cusps in the mammalian molars. From these considerations 

 therefore the form of the dental papilla would appear to be the 

 more reliable criterion. 



With regard to the chemical nature of the materials composing 

 the horny teeth and the scalelets I know of no exact analysis, but 

 it may be pointed out that after the latter have been decalcified 

 with an acid, there yet remains an organic basis probably of the 

 nature of Collagen which is an albuminoid, as is also Keratin. 



The peculiar character of the tooth succession in Petromyzon 

 seems to me to present no very great difficulty. The imbrication 

 of the scales of the Teleostei, each enclosed in its own scale- 



