238 



of the hind legs is such as to render necessary this elongation of 

 the anterior pair of sacral ribs as they are really opposite the 20th 

 vertebra. 



The second variation is much like the one just described although 

 not as symmetrical. On the left side of the 19th and 20th vertebra 

 there are two well defined sacral ribs of equal strength and length. 

 On the right side the 19th vertebra bears a well defined sacral rib 

 while the 20th has a small rudimentary attachment. In both specimens 



the first haemal spine was 

 on the 23rd vertebra. These 

 variations were not detected 

 soon enough to enable one 

 to work out the distribution 

 of the nerves which is to be 

 regreted. 



The previous writers 

 cited have offered various 

 explanations for what is 

 teremed by Bateson as fore- 

 ward and backward ho- 

 moeosis, the migration of 

 the pelvis. Assuming that 

 there is no such thing as a 

 migration of the pelvis, 

 intercalation or excalation, 

 the adding or subtracting of 

 a vertebra anterior to the 

 sacral vertebra is a second 

 suggestion discussed by these 

 same writers. Bumpus and 

 Waite bring fore ward the 

 idea that the stimulus that 

 initiates the formation of 

 sacral ribs comes fi'om the forming legs which do not always develop 

 in the same primitive segments. "What seems most probable is that 

 in different individuals the girdle may develop at primarily different 

 distances (measured in segments) from the cranium." Waite, pg. 86. 

 The most difficult aspect of the problem has been to explain the 

 asymmetry which is very common. Has their been a one sided shift- 

 ing or a double stimulus to girdle formation applied on one side? If 

 homology is taken as an exact criterion then it is hard to see, why 



Fig. 1. 



