Beet webworm.

H. H. Ross, of the Illinois Natural History Survey, observed the effect of reduced rainfall on populations of saw-flies. From 1928 to the spring of 1930, many species of sawflies were present in large numbers in central Illinois. He made intensive collections in two localities, one a prairie habitat along the railroad right-of-way west of Seymour, the other a wooded flood plain and hillside along Salt Fork Creek south of Oakwood. In the first, he found many species of the sawfly genus Dolerus and made daily catches from early April to the end of June of 50 to 500 specimens. At Oakwood, species of Tenthredo and Macrophya predominated. During May, June, and early July of 1930, excessive drought conditions prevailed throughout the Midwest; there were almost no spring rains. All plants wilted and many died. Insects became extremely scarce. Sawflies completely disappeared near Seymour and Oakwood. Not a single specimen was taken near Seymour until 1934, when one specimen of Dolerus was found. Since then there has been a gradual build-up of sawfly individuals in both localities, although never reaching the peak abundance of 1928 and 1929. All the species recorded in 1928 have finally become reestablished near Seymour. Near Oakwood, on the other hand, five of the species recorded as common in 1928 have not been found since. It is interesting to note that the Seymour species were all at or near the southern edge of the species range, and most of the Oakwood species were at the western edge of their ranges. Other North American records of the Oakwood species indicate that this Oakwood area may have been a small "island" on the periphery of the range. If so, the 1930 drought may have brought about a permanent eastward restriction of the range of the affected species. Thus the extremes in weather conditions during the drought weeks of 1930 profoundly affected the distribution and abundance of the sawflies for several years.

TEMPERATURE AND MOISTURE often act together to limit the numbers or distribution of insects. High temperatures and high humidities encourage the spread of a fungus disease that attacks many species of grasshoppers, for instance, and disease may decimate whole populations, nearly to the exclusion of other limiting factors.

An example of probable temperature and precipitation relationships to the distribution and abundance of an insect is given by J. H. Pepper, of the Montana Agricultural Experiment Station, for the beet webworm. That pest is a Great Plains and inter-montane species and feeds on at least 86 species of plants that belong to 33 families. Many of the host plants are spread far beyond the limits of distribution of the insect, and food plants appear to be of no consequence as limiting factors. But when distribution of the webworm is plotted in reference to temperature and rainfall, we can make some correlations that are too close to be mere coincidence.

Eastern distribution of the beet web-worm follows quite closely the 25-inch isohyetal line. (An isohyetal line connects points of similar precipitation, as an isothermal line connects points of like temperature.) Something inherent in the physiology of the pests will not allow them to invade indefinitely an area with more than about 25 inches of precipitation annually. Likewise the western edge of distribution is correlated with an annual precipitation of a little more than 1 inches, although this is not so clear-cut as is the evidence for the eastern boundary.

To the south the beet webworm distribution seems to relate to an average annual temperature of about 55 , and the insect does not extend beyond southern Kansas, northern Oklahoma, and New Mexico, areas that are traversed by the 55 annual isotherm.

The physiological mechanisms that keep the beet webworm from moving out of its area to the east and west are unknown, but the barrier to the south seems probably to be associated with a lack of frozen soil in the winter, for laboratory studies have shown that the common natural method of breaking the resting stage of the larvae and inducing them to pupate is their subjection to freezing temperatures.

The eastern, western, and southern barriers appear to be established, but according to Pepper "their northern range extends as far north as the country has been settled and their limits in this direction possibly have not been reached."

Limits of range appear to have been established by climate, but weather may affect numbers within the area of normal distribution, for the data show generally that a rainfall of 1 to 2.5 inches in each month from April to September is necessary for the most favorable conditions of development.

ANYONE WHO HAS FELT the attacks of salt-marsh mosquitoes blown inland from breeding marshes knows about the power of flight and the ability of insects to be carried by wind two factors that account for their great range.

The direction of distribution of the hessian fly is that of the prevailing winds during the period when the adults emerge. The San Jose scale spreads far more rapidly with the prevailing winds than against them; the wind carries the nymphs as if they were so many particles of dust.

Wind currents, both lateral and vertical, are as much a part of climate and weather as are precipitation and temperature, and their effects on insects vary from the breaking of the resting period, as in some butterflies, to transportation across the face of the earth. Many persons have studied the air as a disseminator of insects. The observations of P. A. Glick were the first extensive explorations of the atmosphere as a distributive medium.

Glick, an entomologist in the Department of Agriculture, developed special traps, which he installed between the wings of a biplane. He made 1,314 flights at Tallulah, La., and 44 at Tlahualilo, Durango, Mexico. In trapping operations that totaled 1,007 hours, he collected 30,033 specimens of insects and arachnids at elevations of 20 to 15,000 feet. The largest number of specimens in a 10-minute trapping interval was taken in May and the smallest in December and January. He captured 18 different orders of insects, spiders, and mites. Among them were 24 species and 4 genera collected for the first time and new to science. Flies were almost three times as abundant in his collections as any other order. Insects were taken at elevations up to 14,000 feet and a spider at 15,000.

The vertical distribution in terms of the average numbers of insects in each 10-minute trapping effort was:

Those figures indicate that above 1,000 feet the atmospheric fauna is relatively static, but below 1,000 feet the number of insects in the air increases at night.

As to whether the insects at higher levels were alive when captured or whether desiccation and low temperatures had destroyed them, Glick wrote: "There is much evidence to support the conclusion that many of the insects taken in the upper air were alive at the time they were collected. Many specimens were alive when removed from the screens. Among the most interesting of these was one mosquito, Aedes vexans, and a cicadellid, Graphocephala versuta, taken alive at 5,000 feet; a coccinnellid, Coleomelailla floridana, at 6,000 feet; an aphid at 7,000 feet, and a small dermestid larva, Trogoderma sp., at 9,000 feet."

In correlating catch with meteorological conditions, he found that most insects were taken when surface temperatures ranged from 75 to 79 F., when the surface dew point was from 600 to 64 , and when barometric pressures were between 29.85 and 29.89 inches. Most specimens were taken at low elevations when the surface wind velocity was 5 to 6 miles an hour, and fewest during a calm. As might be expected, convection and turbulence are important in populating the atmosphere.

"At an altitude of 200 feet more insects were taken when the air was smooth," he wrote. "At 1,000 feet and up to 5,000 feet more insects were taken when the air was rough or slightly rough. As the air became rougher greater numbers of insects were found proportionately at the higher levels."

He made his largest collections below 5,000 feet at daybreak and at sunset. He got more specimens on moonlight nights than on dark nights.