Shufeldt: Birds from Bermuda. 363 



bones of the extinct species. This remark also applies to the pha- 

 langeal joints of the feet. 



Turning the attention next to the lower or pelvic limh, several of 

 the long bones are present, the number of them being set forth in the 

 various lists above. (Compare the bones of the different species of 

 Puffin us shown on Plate XXVIII.) 



A perfect femur of P. parvus occurs in the collection of the Amer- 

 ican Museum of Natural History; it agrees in all respects with the 

 femur of Audubon's Shearwater, as shown in figs. 102 and 103, even 

 to the remarkable bending or bowing of the continuity of the shaft 

 at the juncture of middle and lower thirds. In length it measures 

 two and a half centimeters. 



Femora of other Shearwaters are well shown in figs. 104 and 105, 

 as well as in 112 and 113. The shafts of these are not bowed quite 

 as much as in the smaller species; while in the Petrels of the genus 

 Mstrclata, whether living or extinct, this bone is not bowed at all, but 

 has a straight shaft as in the extinct Petrel, ^strclata vocifcrans 

 (figs. 97-100). This is an excellent differential character in proving 

 that the " Cahow " was an ^strclata and not a Puffinus, as has long 

 been supposed to be the case. 



The tibiotarsus of Puffiuus parz'us presents all the characters of 

 that bone, as we find them in existing representatives of that genus 

 of birds. These characters are well shown in figs. 1 18-120, as corn- 

 pared with the tibiotarsi of other Shearwaters illustrated on the same 

 plate (XXVIII). One of the most striking characters is the enor- 

 mously developed cnemial proce'ss. Comparatively speaking, this is 

 even larger than we find it to be in the big birds of the group, as 

 Puffinus major and others. ^Moreover, this great cnemial extension is 

 not to be found in Petrels, in so far as I have examined them, cer- 

 tainly not in the extinct " Cahow " or Mstrelata vocifcrans. All have 

 the shaft of the bone very straight, however, and this is likewise the 

 case with such forms as Proccllaria cooki and Pelecanoides urinatrix^ 

 which birds have the cnemial extension of the tibiotarsus reduced as 

 we find it in ^strclata. (See figs. 20 and 24 of Pis. XIX and XX, 

 respectively.) 



As in all small and typical Shearwaters, the farsonictatarsus of 

 Puffiuus parvus was long and comparatively slender (figs. 106 and 

 107, PI. XXVIII), it being relatively shorter and stouter in the 



