August 26, 1910] 



SCIENCE 



21b 



certainly inexcusable in this later and larger 

 work. In regard to Asynacta Foerster and 

 Brachista Haliday no attention lias been paid 

 to Mayr's descriptions of their type species in 

 1904, which gave both genera valid standing 

 (formerly without species) and changed our 

 conception of them. This table then is mere 

 copy work. In the brief treatments of the 

 genera following it, the description of Asyn- 

 acta by Arnold Foerster is merely repeated 

 but under Brachista Haliday two of Ash- 

 mead's species are listed without naming 

 either of them as type; I now know that 

 neither of these species belongs to this genus 

 and that Eulophus exiguus Nees is its type as 

 Mayr has designated. Prestwichia Lubbock 

 is treated slightly more at length and figured 

 (plate 8, figs. 9 and 10), but the figure of the 

 male is apparently wrongly copied from Wil- 

 lem (1896) and the male antennse do not show 

 a ring-joint or are but six-jointed. 



But in the second division of the group 

 things are much worse. In the table of the 

 genera Ashmead is again followed except in a 

 few minor instances, and an exact copy of his 

 table would have saved commitment of at 

 least one grave error. Thus Trichogramma 

 Westwood is made to have an exserted oviposi- 

 tor in this exceptional attempt at nice dis- 

 crimination. Otherwise, though the genera 

 are placed in different sequence by shifting 

 sentences, the rest of the tabulation is prac- 

 tically in the same words as given by Ash- 

 mead and all of the errors of the latter are 

 repeated. We look in vain for Ophioneurus 

 Eatzeburg, for Calleptiles Haliday and for 

 Pterygogramma Perkins ; we are wearied again 

 with the same old mistaken diagnoses of Poro- 

 posa Foerster and Trichogramma Westwood; 

 with the needless enlargement of the charac- 

 terization of Chwtostricha Walker {sic) ; with 

 the persistence of Aprohosca Westwood, and 

 with the confusion of Xanthoatomus Ashmead, 

 a genus without status and a synonym of 

 Trichogramma Westwood if such a thing is 

 possible. Hence of the nine genera given in 

 this table four are erroneously diagnosed, two 

 are synonymous with two of the others and 

 one should hardly be accepted — a large per- 



centage of error for such a small number of 

 genera involved; and at least two others were 

 omitted. 



But even this is not all. It remains for the 

 brief treatments of the genera included in the 

 table to bring out still others. In spite of 

 definite and positive statements to the con- 

 trary witness Poroposa being reared from the 

 larva of a beetle; Ophioneurus a synonym of 

 that genus and of Trichogramma; Calleptiles 

 a synonym of the latter; see Trichogramma 

 flavum Ashmead, T. fraternum Fitch and 

 T. orgyice Fitch parading as members of this 

 family; Ophioneurus signatus Eatzeburg in- 

 cluded within Trichogramma. And wonder is 

 indeed excited when we accidentally find the 

 figure of Trichogramma evanescens Westwood 

 (by the way perhaps the only valid species of 

 the genus as far as I am able to learn) on 

 plate 8, figure 3, which is obviously concocted 

 from the imagination and is the more striking 

 because it does not even agree with the char- 

 acters of the genus given in the table of the 

 genera. Pentarthron minutum (Eiley) is 

 listed three or four times under as many dif- 

 ferent names and of that genus a number of 

 species are omitted, the most conspicuous of 

 which is (Oophthora) Pentarthron semhlidis 

 Aurivillius. Lathromeris lacks its only well- 

 described species, cicadw Howard, and Gen- 

 troiia also lacks one of its two species. But 

 why go farther. We can sum up the treat- 

 ment of this family very well in percentages 

 of error. Of the fourteen genera described 

 up to the time of the appearance of the work 

 and which are now valid, 21.5 per cent, are 

 omitted entirely, 14.2 per cent, are given as 

 synonyms of other genera and of the genera 

 actually given, a total of 14, 21.5 per cent., 

 are synonyms, the same percentage are re- 

 ferred back to the wrong authorities, 43 per 

 cent, are wrongly diagnosed or described and 

 there is at least 75 per cent, of error in the 

 figures of the two genera illustrated. Of the 

 44 species described up to the publication of 

 this work as members of the family, Schmiede- 

 knecht omits 30 per cent.; of the 32 species 

 which were then valid, he omits 42 per cent.; 

 11 per cent, of the 27 species given validity 



