ion J Smith, Natural History of Ambystoma. 'l-\ 



weeks they were kept under observation. At the end of five 

 weeks only the largest specimens would take meat. 



A striking morphological feature common to the early larvae 

 of both jeffersonianum and punctatum, but absent in tigrinum, is 

 the presence of ''balancers" at the sides of the head (see Plate 

 III, Figs. 5, 6, 7, 9, io, n). These balancers are sticky at the 

 tips, and serve to keep the larva erect while resting on a solid 

 surface. About two weeks after hatching the balancers degen- 

 erate and disappear ; by this time the broadening of the head, and 

 growth of the anterior limb rudiments, makes them unnecessary. 

 Hay ('91) cites Hoy ('71 ) to the effect that at the time of hatch- 

 ing the larvae of the "tiger salamander" possess short "holders," 

 which are lost ten days later. This would seem to be a mistake 

 in the identity of the species. The absence of balancers is a char- 

 acteristic sufficient to distinguish the larvae of tigrinum from 

 those of jeffersonianum and punctatum 



The most marked difference between the newly hatched larvae 

 of jeffersonianum and punctatum is in the color pattern. Yellow 

 and black pigment are present in both species, and in A. jefferson- 

 ianum there is a tendency toward the segregation of the black 

 pigment along the sides into a row of 4-6 large black spots, shown 

 but faintly in the photographs (see Plate III, Figs. 6 and 7). 

 These spots vary greatly in distinctness in different specimens, 

 but are always present in a considerable proportion of them, and 

 since they are absent in punctatum, afford a convenient character 

 lor differentiating the species. The spots sometimes persist for 

 a month after hatching. 



The newly-hatched larva of A. Jeffersonianum is slightly 

 stouter in body form than that of A. punctatum. The latter is 

 hatched in a slightly more advanced stage of development. 



A synoptic comparison of the larvae of the two species follows 

 (see also Plate III). The body length after hatching is a char- 

 acter too variable for consideration in this connection. 



