402 W. D. M ATTIIKW VlUHUiKSS IN VEUTKlM^VrK P A T-KOXTOlAXn' 



arohitects thereof to exnmino oarot'ully from time to timo \uio the ;ulo- 

 quaov of these t'ouiulations, to eoiuloiun ami swoop away suoh parts o\' 

 our struetiiro as appear to bo insutheiently supported, iliuisy or outworii. 

 They may have served their purpose in the past as temporary outworks, 

 trial sketches or models, or provisional sea (Void in a- to aid in the erection 

 of our more permanent structures; but they should not bo cmiCusod witii 

 solid and stable additions, nor should they be alUnved to outlive their 

 usefulness. A critical review of our foundations and id* their recent 

 extension is the foremost and most impt)rtant matter before us. 



In the early days of paleontology fossil vertebrates wore knmvn from 

 few and mostly very fra^inontary specimens. Our concepts of extinct 

 animals were built up from the study and correlation of numerous frag- 

 ments, supplemented largely by the analogy of living relatives of the 

 extinct animals. The correlations Avere sometimes incorrect, the anal- 

 ogies M'ere always inexact and often misleading. Of the theories and 

 conclusions based by our predecessors on these relatively scanty fomula- 

 tions, some have been swept away and forgotten, some have been moilitled 

 in varying degree, some have boon confirmed and vindicated by subse- 

 quent discovery. 



The more intensive collecting of recent years, and especially the tech- 

 nique devised by Hatcher and Wortman for the purpose of preserving the 

 whole of a fossil skull or skeleton, have brought in }ear by year a larger 

 proportion of complete specimens of fossil vertebrates. The leading- 

 American museums are today peculiarly rich in complete ami well ]u"e- 

 served material, and the more progressive museums of Europe have like- 

 wise adopted these methods, greatly to the improvement of their collec- 

 tions. 



It is difficult to find any basis for a quantitative estimate of the in- 

 crease in our collections, or even of any particular portion of them. So 

 far as the American Museum collections go, the Cope Collection, gathered 

 between 18T2 and 1896, covers about 25 per cent of the catalogue num- 

 bers, but is not in reality over 10 per cent of the collections in this depart- 

 ment, as in former years many specimens were separately catalogued that 

 would not now be considered worth individual record. The other 90 per 

 cent wa5 gathered during the last thirty years, and progressively more 

 during the later decades. Perhaps it would be fair to say that 20 per 

 cent was gathered from 1892-1902, 30 per cent from 1902-1912, and -tO 

 per cent from 1912-1922. Other institutions would have proportions 

 different from these. Probably in Yale University or the National Mu- 

 seum the proportion of material collected and prepared over thirty years 



