Analyses of Books. [September, 
water during the earlier stages of their existence. The lower 
orders of the Hetero-metabola, the Orthoptera and Neuroptera, 
were much more numerous during the Palaeozoic period than the 
higher orders, the Coleoptera and Hemiptera. Almost all the 
Palaeozoic Orthoptera belong to the non-saltatorial families ; 
they are Blattidae, of low organisation. The true Neuroptera 
were also in those times far less numerous than the Pseudo- 
Neuroptera, whose organisation is less perfect. The. general 
type of the structure of the wings of inserts has remained the 
same since the remotest times. Excepting two Coleopterous and 
one Orthopterous species, it maybe maintained that the anterior 
wings of Palaeozoic insedts were similar to the posterior or mem- 
branaceous pair; a difference only appears in the Mesozoic 
period. The venation in types of insedts otherwise distindt was 
more uniform than it is at present. The Devonian, and even the 
Silurian, will probably be found to contain the remains of insedts 
of a type still more generalised than has yet occurred in the 
Palaeozoic strata. Most of the primeval insedts were large, and 
many of them gigantic, and the American and European forms 
are strikingly similar. 
The Food of Birds. By Prof. S. A. Forbes. (“ Transadtions 
of the Illinois State Horticultural Society.”) 
This memoir, which is abstracted and commented upon in that 
most ably-condudted journal the “ American Naturalist,” bears 
so forcibly upon a subject to which we called the attention of our 
readers in our April number that we feel bound to give it our 
consideration. 
The utility of the small birds to mankind, in America certainly, 
and probably in England also, is far greater than we imagined. 
In Illinois there are estimated to be, during the summer half- 
year, three birds to an acre. The author considers that at least 
two-thirds of the food of birds consists of insedts, and that this 
insedt-food will average at the lowest reasonable estimate twenty 
insedts or insedts’ eggs daily for each individual of these two- 
thirds, giving a yearly total of 7200 specimens per acre, or, for 
the State of Illinois, 250,000,000,000, a number which, placed 
one to each square inch of surface, would cover 40,000 acres. A 
careful estimate of the average number of insedts per square 
yard, in this State, gives at the outside 10,000 per acre. Here 
we may remark that anything like exadtitude is difficult. In 
seasons and districts where aphides, midges, sand-flies, &c,, are 
abundant, this estimate seems to us too low. 
To return : if the operations of the birds were suspended, the 
increase of the insedls would be accelerated by about 70 per 
