Karl Pearson and Adelaide G. Davtn 
391 
In Tiliqua scincoides, Helodenna suspectum (the Gila monster), Amieva surina- 
mensis and Callopestes sp. we were only able to note the parafibular sesamoid. 
In Amhlyrhynchus cristatus only lateral posterior and anterior lunulae were 
visible. 
In Physignuthus lesueuri thei-e were all four lunnlae. In Uromastix spinipes 
there was a small lateral and large mesial posterior lunulae and possibly very small 
lateral and mesial anterior lunulae. 
In Jacare punctata we found what was possibly a mesial posterior lunula, but 
this was doubtful. In no mounted specimens of crocodiles or alligators of a large 
size have we found parafibular sesamoid or lumdae*. 
(23) Amphibia. 
Among the Batrachians we dissected two species of frogs without discovering 
any sesamoids at all in the knee-joint. We examined a large number of skeletons 
with the same result. There were, however, two possible exceptions. In Lepto- 
dactylus pentadactylus we believe that a mesial posterior lunula may exist, and in 
Docidophryne gigantea with more doubt a lateral posterior lunula. 
Finally we may note that we have dissected specimens of Trionyx gangetica and 
Testudo graeca and found no ossifications at all in the knee-joint, even the patella 
being cartilaginous. 
It will thus be seen that the Amphibia and Reptilia present a great variety of 
sesamoidal conditions in the knee-joint. We may start from no ossifications at all, 
even the patella being cartilaginous; we may have a single anterior sagittal lunula'f, 
or again two anterior and two posterior lunulae ; or we may find in Lacertilia all 
four lunulae associated with a parafibula. 
We have no real evidence of any lateral fabella in this order. The origin of this 
must be sought in the break-up of the parafibula into cyamella and lateral fabella 
in a stage between marsupial and primitive placental. One point appears to us 
clear, that the appearance of lunulae in the higher orders is a vestige of a reptilian 
ancestry, even in the case where lunulae appear in the Primates (from Lemuroids 
to Man). Further the appearances of cyamella and lateral fabella are vestiges (even 
in Man) of an ancestry of primitive ornithodelphian character which had a para- 
fibula, or possibly a fibular crest. 
The origin of the mesial fabella seems to us shrouded in far greater mystery; it 
is always of secondary importance, often merely cartilaginous, still more often absent; 
it is not a "Knotenpunkt" of important muscle attachments like the lateral fabella. 
The cnemial crest of the tibia is relatively rare and has been very rarely indeed 
* In a mounted specimen of Crocodilus americanus we noted what looked like a very tiny external 
fabdla I 
t The possibility of an anterior sagittal lunula being the source of a patella is possibly worth bearing 
in mind. 
