THYSANOPTERA AND ORTHOPTERA. 221 



From these observations there seems httle doubt that the insects 

 are more active in the middle or latter half of the afternoon. Table I 

 shows little except the comparative sluggishness of the grasshoppers 

 in the early morning. Table II is of particular interest because of the 

 nearly uniform temperature, notwithstanding which there is a notice- 

 ably greater activity during the morning hours. The fact that the 

 two lowest counts, however, were obtained at the two lowest tempera- 

 tures (see 7:35 a. m. and 2:00 p. m.) gives room for supposing that 

 temperature is partly responsible for the difference in activity. This 

 is not borne out by Table III, unless it be further supposed that in- 

 crease of temperature is accompanied by increase of activity up to a 

 temperature of about 24° C, and that further increase of temperature 

 is productive of a decrease of activity. This is entirely possible, but 

 I lack data to decide positively. 



The results should be of interest not merely to a collector, but to 

 those who haA^e recourse to mechanical means of killing the insects as 

 an economic measure. 



ECOLOGY. 



An attempt was made, after collecting the season's data, to fit these 

 into the scheme of ecology proposed by Morse (1904, p. 14) for the 

 Acridiidae. This scheme was borrowed at least in part from the 

 botanists and transferred bodily to the insects. When my data for 

 the various families were studied, it soon became apparent that the 

 scheme w^as not readily applicable to other families than the Acridi- 

 idae; it was clearly not practicable to classify many of the specimens 

 taken as hygrophile or xerophile, sylvan or campestrian, humicolous, 

 arenicolous, or saxicolous. With the Acridiidae, the case was some- 

 what better; but even here there were many contradictions. The same 

 species was often found in very chfferent locahties, for example in 

 marshy spots or in the driest fields, on humous soil or on nearly bare 

 sand. This may be attributable to the nature of the region. There 

 were no large unbroken areas of uniform character, so that every loca- 

 tion was practically on the edge of some habitat of a different kind. 

 Contradictions in habitat may thus be accounted for by chance migra- 

 tions, without necessarily supposing the insect to have no preference 

 for one location. Under such circumstances, the only way to arrive 

 at a satisfactory knowledge of the ecology is to collect immense num- 

 bers and depend upon majorities to decide the usual habitat. Such 

 large numbers were not collected. It appears that the best way to 

 arrive at a consistent scheme is to collect thoroughly over extensive 

 uniform areas. Until this is done, it is of little value to work out an 



