498 PROCEEDINGS OF THE ACADEMY OF [1898. 



ing several anatomical peculiarities, this genus is one containing 

 several species of ducks, and ducks not so very far removed from 

 either the Teals, or the Mallard, or perhaps Spatula. There is very 

 little Goose, and still less Swan in the morphology of Dendroeygna, 

 and for what reason the genus has been placed between Philacte and 

 Olor in the Chech-List it is difficult for me to understand. In 

 the first edition of his Manual Mr. Ridgway places Dendroeygna, 

 the last genus in the duck-series where the synopses of characters of 

 the Anseres are set forth, while in the part devoted to the diagnoses 

 of species and genera, these Tree-ducks are placed between the 

 Swans and the Geese as in the Check-List. They have as I have 

 just said some peculiar characters about them, and of these, one of 

 the most interesting is the fact that they have complete bony rings 

 surrounding the orbits, as is the case in several genera of parrots 

 and some other birds. So far as I am aware it is the only genus of 

 ducks that presents this character, — indeed, the only anserine bird 

 that has it. 



Coming to the Cranes and Rails we meet with an interesting form 

 in Aramus giganteus. During the past few years I have compared 

 the skeletons of several hundred species of birds, and written out the 

 osteology of nearly every genus in this country, and among all these 

 have been included the entire Crane and Rail group with all the 

 North American birds in any way related to it. In this manner have 

 Grus, Aramus, Rallus, Porzana, Orex, Ionornis, Gallinula, Fulica, 

 and others been dealt with, and their skeletal characters arrayed in 

 tabular form in great detail. Without entering upon the general 

 taxonomy of this group, it is an interesting fact, that in so far as the 

 skeletal characters are concerned, Aramus presents two for every 

 one in favor of its affinity with Grus as compared with Rallus, yet 

 in nearly all avian classifications we find this bird arrayed with the 

 typical Rails. Four years ago I published in England, an abstract 

 in which was incorporated some of the facts here stated, with part 

 of a scheme for the classification of this group. Since then I have 

 examined a number of forms at that time not available, and although 

 they have not materially altered my original views, some changes 

 will necessarily have to be made in order to include these facts which 

 have since come before me. 



Of recent years nothing has come to my notice that seems likely 

 to again check the now growing opinion that the Woodpeckers, as 

 another assemblage of birds, see their nearest relatives in the Passeres, 



