ORNITHOLOGICAL CLUB. 65 



of immigrants may have been maintained from the earliest times. 

 It must be remembered, however, that no precise order of immigra- 

 tion can be laid down, even approximately ; for one term in the 

 proposition is an absolutely unknown quantity. We know nothing 

 of the rate of change that has taken place in any one species. 

 Outward circumstances may have acted on one species, so as to 

 leave it little changed in a given time, whilst in the same time an- 

 other species may have assumed distinctive generic characters. 

 Viewing the very peculiar physical characters of these islands when 

 contrasted with the neighboring American shores, it would seem 

 reasonable that the rate of change demanded of an immigrant 

 species would be high ; consequently the origin of the islands need 

 not be dated back to a more distant period than seems indicated by 

 their volcanic origin." 



Considered in connection with the subject discussed above, the 

 birds of Guadalupe are of extreme interest, since they apparently 

 represent a transition stage through which those of the Galapagos 

 once undoubtedly passed. Nothing, unfortunately, is known to the 

 writer as to the geological structure of Guadalupe ; the character of 

 the modifications presented in its birds, however, point strongly to its 

 volcanic origin, and render it extremely probable that the upheaval 

 took place at a more recent date than that of the Galapagos. The 

 earliest immigrants to this island were probably the ancestors of 

 Polyhorus lutosus, which has become completely differentiated in 

 plumage but not perceptibly altered in the details of structure,* and 

 those of Carpodacus amplus, whose modifications of external struc- 



* The case of tliis species presents a very curious problem. Its origin from 

 P. clieriway, the only species now inhabiting Middle America, and even north- 

 ern South Ameiica, can scarcely be doubted ; but the modifications which the 

 Guadalupe species has undergone tend toward the distinguishing characters of 

 the South American form {P. tharus). The two continental representatives of 

 this genus have undoubtedly had a common origin, the differences between them 

 coming under the scope of ordinary geographical laws of variation in this 

 family, as at present understood. The differentiation of the Guadalupe form is 

 of a most remarkable kind, however, being apparently a partial reversion to the 

 features of the Southern form ; but some of the characters which distinguish 

 the latter from its Northern analogues are even greatly exaggerated in this North- 

 ern insular form ! In this instance, then, the differentiation has been a kind 

 of retrocession, with no change in details of structure, while in all the other 

 forms of the island the differentiation has been of the opposite kind, affecting 

 the proportions more than the colors. 



