Insects To Control a Weed

James K. Holloway, C. B. Huffaker.

Over many square miles of western range lands millions of pea-sized, bright, metallic-colored beetles are destroying a common weed that for years has caused huge economic losses.

The weed, Hypericum perforatum, has a number of common names. In California it is called Klamath weed because it was first reported, about 1900, in northern California in the vicinity of the Klamath River. In many of the Western States it is sometimes referred to as goat weed. The recognized common name in Europe, original home of the weed, is St. Johnswort because, according to legend, it blooms on June 24, the day of St. John the Baptist.

It has invaded extensive temperate regions throughout the world. It is considered a noxious weed in the range lands of Australia, New Zealand, Canada, and the United States. The infested areas in California are estimated at 400,000 acres. Oregon, Washington, Idaho, Nevada, and Montana also have many thousands of infested acres.

Klamath weed causes losses by displacing desirable range plants. It is poisonous to livestock, but death as a result is rare. Animals that eat much of it become scabby, sore-mouthed, and unthrifty. It causes the white parts of the skin to become photosensitive and, when exposed to sunlight, blisters form on the unpigmented skin areas. Cattle are more sensitive to it than sheep.

In some localities in California the grasses dry rapidly in the spring, and the most abundant remaining green plant is Klamath weed. In those localities cattle are usually moved from the ranges before the condition arises. But on occasion unavoidable delays do occur, and the animals may then consume damaging quantities of the weed. The ingested plant causes cattle to become irritable so that they are difficult to corral; sometimes it is almost impossible to load them into trucks and it may be necessary to confine them and give them other feed for a day or two, until the effects of feeding on the weed wear off.

Many attempts have been made to control the weed, a perennial, with chemicals borax, 2,4-D, and others. But the materials are expensive and the land to be treated is mostly extensive and inaccessible.

Control of the weed by insects has been under consideration for several years. The general method was successful in other countries, notably Australia, but its use in this country is a recent development.

The Commonwealth of Australia first began a search for insect enemies of St. Johnswort in 1920 in England. Early in 1935, after the insects imported from Britain proved apparently unsuccessful in Australia, the search was transferred to southern France. The early work in Europe comprised tests by starvation and breeding of many insects on 42 species of economic plants, representing fig botanical families, to determine whether the insects could feed and breed on them.

At the satisfactory conclusion of the tests in Europe, the species that had shown neither feeding nor reproduction upon the test plants in Europe were shipped to Australia. Before they could be liberated, however, additional tests had to be made on plants that had not been tested in Europe.

About 8 years after two species of the leaf-feeding beetles, Chrysolina, were released in Australia, encouraging results were reported. Subsequently men at the University of California who had watched the experiments with great interest were authorized by the Department of Agriculture to import Chrysolina hyperici, C. gemellata, and a root borer, Agrilus hyperici. The stipulation was made that feeding tests be made on sugar beet, flax, hemp, sweetpotato, tobacco, and cotton.

A project for the importation, testing, and colonization of the three species was then set up by the Bureau of Entomology and Plant Quarantine and the University of California.

The war made it impossible to collect the insects in Europe. It was learned, though, that abundant material was available in Australia and would be transported to California by the United States Army Air Transport Command. The Australian Council for Scientific and Industrial Research offered to collect and ship the material.

Leaders in those activities were Frank Wilson and Harry S. Smith. Wilson was sent to southern Europe in 1935 when the first shipments of the natural enemies of Hypericum per oratum from England to Australia failed to progress satisfactorily. Wilson had been associated with the work in England and he continued it in France until 1940. Professor Smith, who was head of the division of biological control in the University of California until his retirement in 1951, is regarded as one of the world's foremost proponents of biological control. Ever since the biological control of Klamath weed was advocated by Dr. R. J. Tillyard of Australia in 1926 he followed the developments. In correspondence with Dr. A. J. Nicholson of Australia in 1944, Smith found that the biological control was beginning to make progress. He then took the steps that led to the project between the University of California and the Department of Agriculture. When the importations began, James K. Holloway was put in charge of the investigations.

THE FIRST IMPORTATIONS were made in October 1944. The initial problem was to adjust the life cycles so that they would be in phase with the seasons of the Northern Hemisphere. The specimens of the root borer (Agrilus hyperici) were received as mature larvae in roots. Some of them were retarded in cold storage, but others were forced to emerge upon arrival. Neither method proved satisfactory, and further importations were curtailed until the work in Europe could be resumed.

The two species of Chrysolina were Occasionally shipped as mature larvae, which would emerge as adults upon arrival, feed, and enter summer dormancy about 3 weeks later. Most of the shipments, however, consisted of summer-dormant adults. Either way, the problem was to bring the adults out of aestivation into the egg-laying phase. By subjecting the adults to fine sprays of water each day, a state similar to normal winter moisture conditions was reproduced in the laboratory, and the beetles came out of aestivation, mated, and began producing fertile eggs within 2 to 3 weeks.

During the first year of importations, enough C. hyperici were received to conduct the feeding tests. The tests were completed in May 1945. No feeding had taken place on any of the test plants, and four colonies were released late in the season.

In January 1946 the feeding tests with C. gemellata were completed. Permission was obtained to release 13,650 adults that were being retained in quarantine. They were divided into two colonies of 5,000 each, one colony of 2,000, and one of 1,650.

The experimental releases of both species were made in the Coastal Range, northern Sacramento River Valley, and the Sierra foothills localities considered representative of the grazing areas in which the weed occurs in California.

A total of 330,000 adults of C. hyperici, shipped from Australia, was released in 1947 at 66 sites in 15 counties of California. Two experimental colonies of 5,000 each were released in Oregon through the cooperation of the Oregon Agricultural Experiment Station.

Both species of Chrysolina were well established by 1948, and we did not need to import more. Two of the original releases of C. gemellata had shown a remarkable increase, and from them we collected 212,000 adult beetles, which we placed in 52 new locations in 16 counties of California. Three other releases were made in Oregon and two in Idaho. Initial releases of C. hyperici were made in Washington, Idaho, and Montana in 1948 through the cooperation of the State experiment stations. The Forest Service joined the project in 1949 in Oregon, Washington, Idaho, and Montana. In May of that year 140,000 adult C. hyperici were collected and shipped to those areas in units of 5,000 each, making 28 new locations in the Northwest.

C. gemellata was so numerous and widespread in California by 1950 that redistribution became a local problem. Perhaps 3 million adult beetles were collected and redistributed in May 1950.

The success of C. hyperici has been limited. It has become established in other localities, but an increase comparable to that of C. gemellata has been restricted mainly to the coastal mountains in California.

The effectiveness of the leaf-feeding beetles in controlling Klamath weed is associated with their life cycles and their weed host. The balance between the propagative ability of the weed and that of its insect enemy is determined by factors of soil condition, climate, and the influences of interrelated plants and animals.

The two species of weed-feeding beetles differ slightly in their environmental requirements. Yet that small difference means that one species reproduces abundantly and the other's reproduction is curtailed under California conditions.

C. gemellata starts reproducing quickly when the fall rains come. Consequently its progeny have enough time to make the necessary growth before the dry season arrives in late spring and early summer conditions that are hazardous to pupal development. Also, the egg-laying period is longer, and greater numbers of eggs are deposited.

But C. hyperici reacts slowly to moist conditions in the fall and under average California conditions deposits most of its eggs so late that there is insufficient time to complete the necessary phases of development before dry weather sets in.

The life history of the more successful species is attuned to the phases of weed growth and to local climatic conditions. The adult beetles issue from their pupal cells just beneath the surface of the soil in April and early May. They feed voraciously during May and June on the foliage of the plants, which then are flowering. By late June and early July the beetles have completed preparation, by feeding and sunning, for their summer sleep. This inactive, dry-season stage is spent beneath debris, under small stones, and in crevices of the soil.

The beetles spend 4 to 6 months in this inactive condition without food or water. During this period the weed, too, enters a relatively dormant phase. It develops and ripens its seed crop but drops most of its leaves and becomes hard and woody.

The larvae feed actively in warm periods in winter and spring. Their intensive feeding keeps the plant stripped of leaves over a long period when its food reserves are at a low ebb. Thus the root system and the plant die of starvation. Adult feeding, voracious as it is, does not last long enough to produce wholesale death of the plants without the previous feeding by the larvae.