410 



SCIENCE. 



[N. S. Vol. III. No. 63. 



peck instinctively and did not offer to take food 

 spread before them. The natives seemed well 

 aware of this peculiarity, and in the JDarticular 

 instance recorded a native induced the young 

 birds to peck by tapping on the ground with a 

 pencil near the food. They seemed attracted 

 by the sound and movement, and were thus in- 

 duced to peck at the food. F. A. Lucas. 



SCIENTIFIC LITER ATUBE. 



bkongniart's paleozoic insects. 



Mecherches pour servir a I'histoire des Insectes 



fossilesdes temps primaires, precedees d'une 



etude sur la nervation des ailes des Insectes. 



Saint Etienne, 1893. 2 v. 4°. Text, 493 pp.; 



Atlas, 44 pp., 37 folding plates. 



These volumes, which are primarily devoted 

 to the carboniferous insects of Commentry, 

 France, form the most important work that has 

 ever been published on paleozoic insects. Our 

 knowledge of the older hexapods has hereto- 

 fore been obtained piecemeal, and generally by 

 exceedingly fragmentary researches; while here 

 we are introduced at once to a wealth of ma- 

 terial equalling, if it does not surpass, all pre- 

 vious knowledge of paleozoic insects. Mr. 

 Brongniart had indeed published a few of his 

 interesting finds in previous minor papers and 

 had given also a summary account of the Com- 

 mentiy fauna in a brochure in 1885 ; but as the 

 latter contained almost no details, and was 

 merely a sketch of his classification (here modi- 

 fied in a few particulars), it had slight value ex- 

 cept as a forecast of what is now realized. 



Cockroaches form in all Carboniferous de- 

 posits the major part of the insect remains, and 

 many hundreds of specimens have been ob- 

 tained at Commentry. Leaving these out of 

 account because reserved by the author for 

 future publication (a few figures only without 

 descriptions being given), the fauna of Commen- 

 try consists, according to Brongniart, of Neu- 

 roptera, Orthoptera and Homoptera; these he 

 divides into 12 families or larger groups, ten of 

 which are regarded as extinct, and they include 

 48 genera and 97 species, a number of species 

 just about double that of the previously known 

 European Carboniferous hexapods, exclusive of 

 course of cockroaches. 



The variety, novelty and striking character 

 of the forms revealed is as interesting as their 

 number. No one of them, indeed, can be re- 

 garded as extraordinary as Eugereon ; but we 

 are introduced to long-winged giants regarded 

 by Brongniart as the precursors of the Odonata, 

 but which in spread of wings make our largest 

 dragon-flies appear as pigmies ; one, Meganeura, 

 has a spread of considerably more than two 

 feet, and one specimen of this, which I have 

 had the good fortune to see, is so well pre- 

 served that four nearly perfect and fully ex- 

 panded wings are in place attached to the 

 thorax ; others have saltatorial hind legs as 

 fully developed as in our existing Locusta- 

 rians, but with very different wing neuration. 

 Thyspanura (before known fossil only from the 

 Tertiary) are indicated — unfortunately not fig- 

 ured — which have but a single caudal seta ; 

 more than fifty specimens of this have been 

 unearthed. Insects are found with a broad 

 lobate expansion on either side of the pro- 

 thorax, recalling some living Mantidaj (Chcera- 

 dodis, etc.), but which, being filled with apparent 

 nervures, Brongniart regards with too great 

 confidence as prothoracic wings. Others, and 

 these include a variety of types, have lobate 

 appendages at the sides of all the abdominal 

 segments, like the branchial gills of the larvae 

 of some existing Neuroptera, persistent through 

 life in Pteronarcys. There are also gigantic 

 Mayflies, and Neuroptera of large size with 

 caudal setse more than six inches long. And, 

 finally, we may mention undoubted cockroaches 

 which show a straight, slender, Locustarian- 

 like ovipositor half as long as the abdomen, an 

 additional and striking difference to distinguish 

 them from modern cockroaches. 



Brongniart begins his work with a somewhat 

 detailed historical review of discoveries in the 

 field of paleozoic insects, with an appended 

 bibliography, and follows it by an extended 

 study of the neuration of existing Neuroptera, 

 Orthoptera and Fulgoridte (180 pp.), as a basis 

 for his attempt to classify the Carboniferous 

 forms; 12 of the plates are also given to the 

 illustration of the wings of modern insects. 

 In this study he follows with some modifica- 

 tions the guidance of Redtenbacher, apparently 

 unaware of some later studies on the subject, 



