OF WASHINGTON. 183 



Great, therefore, was the writer's delight when, on a short 

 visit to the Academy last fall, he opened an old-fashioned double 

 box, which had turned up in the interval between his visits, and 

 found one side filled with what, he at once realized, was the 

 nearly complete set of Clemens' original types. 



Pinned as they were on the short English pins, many of them 

 touched the cork with their wings and several were consequently 

 more or less damaged ; but considering their old age and their 

 precarious method of preservation in a box which was neither 

 dust nor insect proof, it is rather remarkable that they had not all 

 been destroyed. However, all were in recognizable and useful 

 condition, and some of them in a perfect state of preservation. 



On this visit the writer was unable, from lack of time, to do 

 more than merely satisfy himself of the genuineness of this col 

 lection of types ; but shortly afterwards he had the opportunity, 

 through the liberality of the U. S. Department of Agriculture, 

 and on the invitation of the Academy in Philadelphia, to spend 

 two weeks in the study and resurrection of this important collec 

 tion, which he regards it a privilege to have been able to restore 

 to the dignity due the founder of this branch of science in this 

 country. 



That these specimens truly are Clemens' authentic types is 

 proven by Cresson's statement and by the list in Clemens' own 

 hand, corresponding to the numbers on the specimens. It is 

 further verified by several instances, where Clemens, in his de 

 scription, mentions accidental peculiarities of the specimens before 

 him, which are found to be present in the corresponding type. 



The numbers on the first 124 specimens correspond, with a 

 few easily explained exceptions, with the chronological order of 

 Clemens' descriptions, and these were found in the same order in 

 Clemens' handwritten list. The numbers on the rest of the types 

 were seemingly without order and many intervals occur. On 

 the whole, however, these also were found to correspond rela 

 tively to the chronological order, and the intervals can be ac 

 counted for by intervening types of other groups, though several 

 unexplainable deviations occur. By careful verification of each 

 species through the description it was not difficult for one some 

 what familiar with the different forms to apply each type to its 

 proper name, taking as starting points species already well known 

 and working forwards or backwards according to the numbers. 



The studies resulted in the identification of the types of all but 

 eight of Clemens' 200 species. Five of these eight have been 

 identified with certainty from Clemens' descriptions, leaving only 

 three species unknown at present. 



The types are now properly mounted on small corks pinned 

 with stout pins in four Schmitt boxes. They can now be taken 

 out and examined without unnecessary risk to the specimens, and 

 each of them is correctly labeled. 



