128 M. de Quatrefages on 



materials soon enabled him to recognize five distinct species, 

 wliich he united into the genus Dinornis. 



Subsequently this number gradually increased to thirteen, 

 and there have been found in these representatives of an 

 extinct fauna more and more marked differential characters. 

 This is so much the case that Dr. Julius Haast, the eminent 

 New-Zealand geologist, has thought it right to form of them 

 four genera, themselves divided into two groups or families *. 



It is easy to see that these pala^ontological discoveries con- 

 firm the observations which I have just been making and 

 bring New Zealand under the general rule. This austral 

 country has never produced Mammalia. To make up for 

 this the type of the short-winged birds has been developed 

 there with an abundance and a variety of secondary types 

 such as we meet witli nowhere else. There is complete 

 accordance between its fossil and its recent faunas, and these 

 faunas, precisely by the exceptional character which is 

 common to them, furnish another proof of the universality of 

 the laws which everywhere bind together the past and the 

 present of the animal world f. 



lowed out his researches chiefly bj^ the aid of the materials sent by Mr. 

 W. Mautell. The results have appeared iu the ' Transactions of the 

 Zoological Society ' for 1844 and following years. 



* The following is Dr. Haast's classiiication, which, however, only 

 includes eleven species : — I. Family Dinornithid^ : genus Dinornis, in- 

 cluding D. maximus, 1). rohustns, I), inyens, D. struthioides, D. (/racilis ; 

 genus Meionornis, including M. casuarinus, M. dicUformis. II. Family 

 Palapterygid^ : genus Pcdapteryx, including P. elephantojnis, P. cras- 

 stis; genus Eiiryapteryx, including E, gravis, E. rheides (Proc. Philos, 

 Inst, of Canterbury, March 1874 ; Addi-ess by J. Haast, president ; Trans- 

 actions, &c., vol. vi. p. 426). Dr. Haast, from considerations drawn 

 especially from size, seems disposed to think that he has himself united 

 under the name of Meionornis casuarinus two species which will have to 

 be distinguished hereafter. He makes analogous remarks with regard to 

 Palapteyyx elephantopus (p. 429). Prof. Hutton, Director of the Museum 

 at Otago, has criticized Haast's classification, and denied some of the 

 facts relied upon by his confrere. With Owen, he thinks that the Moas 

 form only a single natural family, that of the Dinornithidse (' Transactions 

 &c., vol. ix. p. 363). Owen and M. A. Edwards only admit the two 

 genera Dinornis and Pcdapteryx, the former tridactyle, the latter having 

 a fourth digit, which is short and directed backward. 



\ The preceding observations apply not only to the history of the New- 

 Zealand fauna, but affect the history of man himself. By themselves 

 they suffice to refute a theory recently put forward by M. P. A. Lesson 

 in a book in other respects filled with important facts and documents, of 

 which three volumes out of four have appeared, namely ' Les Polynesiens, 

 leur origine, leurs migrations et leur langage ' (Paris, 1882). The author 

 assumes that the whole of Polynesia, Tahiti, the Sandwieli Islands, the 

 Samoa and Tonga Islands. &c. has been peopled by means of migrations j 



