and autumn egg surveys. If control begins before the hatching of the major species is completed, it may have to be repeated. If it is continued after egg laying is well under way, it has little effect on preventing grasshopper damage the following year. If the fall egg surveys are made before peak egg laying is completed, a false picture of the next year's potential infestation is obtained. O. L. Barnes used the data from study areas in Arizona in making recommendations for the timing of grasshopper surveys in that State.

In Montana and North Dakota, where the lesser migratory and two-striped grasshoppers are the dominant economic species, eggs were found most heavily concentrated in roadsides, fence rows, idle land, and small grains. Few were found in row crops or fallow land. Knowledge of where eggs are laid is indispensible in making egg surveys. Weedy roadsides and fence rows harbored more grasshoppers and contained more than twice as many eggs as similar places that were covered with solid stands of grass. E. G. Davis used that information as the basis for recommending the regrassing of weedy roadsides and fence rows to reduce grasshopper damage in nearby crops.

One of the study areas in Montana was in the Centennial Valley in Beaverhead County. The clear-winged grasshopper, the dominant species there, in some years completely destroyed the hay crop. The studies disclosed that this species concentrated its eggs in the roots and crowns of white bunchgrass associations (Poa and Puccinellia), which occur in well-defined patches throughout the native hay lands, constitute only a small fraction of the total hay acreage, and are easily recognized by their light color. Landowners were asked to bait the white bunchgrass whenever they saw grasshoppers in more than ordinary numbers. No outbreaks have occurred since.

Data from Montana and North Dakota provide estimates of what happens to the yearly potential grasshopper population. Females of the more important species lay approximately 200 eggs. Males and females are about equal in numbers. If the same level of population from one year to another is maintained, 198 (99 percent) of the eggs, young grasshoppers, or adults must perish sometime between the completion of egg laying and the beginning of egg deposition the following year. During the to-year period, approximately 20 percent of the eggs were destroyed by predators, 60 percent of the young grasshoppers died during or shortly after the hatching period, and disease and parasites killed 5 percent of the older nymphs and adults.

Weather conditions during and following the hatching period may start or end outbreaks. If it is warm and dry, the death rate may decline below the 60-percent average and allow grasshoppers to develop in outbreak numbers. If it is cold and wet, mortality may increase to such an extent that few survive.

OUR KNOWLEDGE of grasshoppers on crops has been greatly furthered by seasonal studies in typical agricultural areas. R. L. Shotwell and I have described the 1931 outbreak of the two-striped and differential grasshoppers in South Dakota and northeastern Nebraska, which destroyed 75 percent of the crops on 17,000 square miles and 25 percent on an additional 13,000 square miles. Shotwell has published a bulletin on the histories and habits of grasshoppers that attack crops in the Northern Great Plains. C. C. Wilson reported his observations on the devastating grasshopper, an important economic species in California.

Surveys of numbers of adults and eggs have been conducted by State and Federal entomologists for many years and are of importance in planning control operations. Surveys have evolved from reconnaissance (which merely delineated areas of greatest abundance) to the present standardized type, in which infestations are classified according to the number of adults per square yard and the number of egg pods per square foot of soil.

One of the problems in connection with annual grasshopper egg surveys on the permanent study areas, as well as in State surveys, was the number of samples needed to rate accurately the fields and field margins and the number of stops needed per section, county, or other area unit. Special surveys designed to answer these questions were conducted in Montana in 1939 and 1940 and in South Dakota in 1942. The results of the studies were published by Davis and F. M. Wadley. They concluded that five 1-square-foot samples in each field and two similar samples in its margin should be taken and that no fewer than 10 field stops should be made in a county or group of counties.

RANGE GRASSHOPPER studies have been conducted as a major project since 1936. They include yearly observations at 10 permanent range stations in Montana, 3 in Wyoming, and 2 in South Dakota; adult and egg surveys of the western range region extending from Montana and western North Dakota to Wyoming and western Nebraska; and seasonal observations in places where significant range grasshopper activities are in progress.

State and county distribution of 142 species of grasshoppers found on the western range has been mapped. Adult and egg surveys conducted since 1941 have recorded over-all population trends and the dominant species for each year. Egg-laying records, including the number of pods laid and the number of eggs per pod, have been made for 29 species. Egg pods, nymphs, and adults of the more common species have been photographed or described.

Studies of seasonal development in Montana and Wyoming have shown wide differences among the complex of species found in any locality. Some are early; others are intermediate; and the rest are late in hatching, reaching the adult stage, and egg laying. Differences of 4 to 5 weeks have been found between early- and late-developing species. Knowledge of the seasonal development of the dominant range species is essential to the proper timing of control operations and surveys.

We divide range grasshoppers into two groups on the basis of food preferences those that feed on grass and those that feed on fortis (which are range herbs other than grass). Of 40 species under study for three seasons in southeastern Montana, half were grass feeders and half were forb feeders. Those that prefer grasses will starve if restricted to fortis, which they refuse to eat in quantity. Forb feeders starve on grass unless it is tender and succulent. Several grass-feeding species would not eat bran bait. Trachyrhachis kiowa and Opeia obscura ate no bait. Phlibostroma quadrimaculatum, Amphitornus coloradus, and Metator pardalinus ate it sparingly. When those species predominate, baiting is ineffective.

Egg predators and parasites of nymphal and adult range grasshoppers have been investigated. Two species of nemestrinid flies, Trichopsidea (Parasymmictus) clausa and Neorhynchocephalus sackenii, not previously reported as common parasites of grasshoppers in the United States, were discovered in southeastern Montana in 1949. Females lay their eggs in cracks in wooden fence posts and trees and do it fast. One female laid 1,000 eggs on a single post in 15 minutes. Another confined in a pill box laid 4,700 eggs in 7 hours. The eggs hatch in 8 to 10 days. The tiny maggots are scattered by the wind. When a maggot finds a grasshopper, it bores into the abdomen and lives on its contents. Eventually it kills the grasshopper. Nemestrinid maggots differ from the better known sarcophagid parasites in having long breathing tubes, which they attach to the grasshopper's air circulating system to get their own fresh air. Nemestrinid flies are particularly destructive to the important range species Metator pardalinus; 80 percent of the adults of this species were found parasitized in some localities in 1950. The remarkable number of eggs laid by nemestrinid females and the ease of obtaining them suggests the possibility of gathering large quantities for distribution in grasshopper areas where parasites are not abundant.

Many observations have been made to determine the damage done to range vegetation by known numbers of grasshoppers. During drought years, infestations of 20 to 50 per square yard frequently have destroyed 75 to 100 percent of available forage by midsummer. From outdoor cage experiments conducted in range areas for several years we learned that an adult grasshopper of the larger range species ate 30 milligrams of vegetation (dry weight) a day. In earlier laboratory experiments we found that the relatively small, lesser migratory grasshopper ate 24 milligrams a day. A cow requires 20 pounds of vegetation (dry weight) a day. From these data we calculated that 301,395 adult range grasshoppers and a cow would eat equivalent quantities of forage a day. This number of grasshoppers distributed over 1 acre would average 62 per square yard, or 31 per square yard over 2 acres. The fact that such populations are common during outbreaks shows the stiff competition between grasshoppers and livestock for range vegetation. The same data show that infestations of only 7 grasshoppers per square yard on an acre will eat one-tenth as much forage as a cow. Control measures are seldom employed against such numbers of grasshoppers even though they lower the livestock-carrying capacity of the range.

Many long-time residents in western range districts have commented that range grasshopper outbreaks in their neighborhood always started in certain relatively small places and then spread to surrounding land. Fifteen potential outbreak localities in eastern Wyoming were selected for study. They have been under observation since 1941. Yearly population records have been made for each area and for the range adjacent to it.

Those studies showed that trends during low years were not necessarily uniform; that is, some areas were slowly declining while others held their own, and still others built up slightly and declined again. Build-ups tended to accompany an expansion of the infested area involved, while declines accompanied a contraction of infested areas. During 1946 and 1947, gradual increases occurred at about the same rate both within hold-over areas and outside them, but independently of each other. The increases continued in 1948 and 1949, and the hold-over areas with their greater populations were the first to reach damaging levels. When they did, there was a tendency for local migrations and light flights to take place some of them to the outside. Drought tended to stimulate both migrations and flights. It was evident, however, that with a continuation of favorable years, outside populations could, and in some cases did, reach damaging levels without contributions from the hold-over areas. Over-all populations built up through slight to moderate increases of most major species rather than by large increases of a few species.

The results of the studies are disappointing to those who hoped that widespread outbreaks of range grasshoppers could be prevented by prompt use of control methods on a few small areas whenever numbers approached outbreak proportions. They do show that prompt control in hold-over areas is highly desirable, but they also indicate that similar action may be needed in many other places when conditions become unusually favorable for grasshoppers.