From then on the weed suffers progressive destruction of its foliage by the larvae. The fully mature larvae enter the soil to pupate at about the time the plant begins to develop the shoots that become the flower-bearing stalks. The appearance of the adult beetles completes a single cycle, which covers a year.

The beetles can move in effective numbers into new areas, but enough of both sexes must be present to assure fertility of the eggs. About 3 years generally are required to give local control in a remote area where only a few thousand beetles have been released. In the third reproductive year the numbers reach a level at which they can exert controlling pressure on the weed.

Local dispersion is normally accomplished by the crawling adult beetles, which often are seen moving in great numbers across roads or ravines from centers of overpopulation where the weed has been cleared. Dispersion by flight is less commonly observed but occurs under conditions of high beetle density, complete depletion of the food supply, and hot, sunny weather. Colonies have appeared which apparently are the results of single or repeated flights from production centers as far as 3 miles distant.

Through its natural powers of dispersion, plus a supplemental influence from establishment of secondary initial colonies in the area, C. gemellata has now spread and effected general control of the weed over hundreds of square miles in southern Humboldt County, California.

The same species has cleared an open range area in Placer County and moved from that limited infestation through small, isolated patches of weed to points up to 3 miles away in various directions. No additional releases were made anywhere in the area. A second and younger colony in Placer County was located in an area notably unfavorable as to climate, yet it cleared that field of the weed within 3 years.

C. hyperici, though poorly adapted to the California conditions generally, brought under control 4 to 5 square miles in an upland area of Humboldt County, where retention of soil moisture late in the season has operated to its advantage.

Ranchers and farm workers attest to the complete destruction of the weed by the beetles. They have seen how the hungry insects have removed the weed from a large area near Blocksburg, Calif.

The beetles also can locate and destroy small, isolated stands of weeds that were missed in previous years. The appearance of the seedling weeds in cleared fields is common, but so far enough beetles are present in the area to find them. These scattered reinfestation spots and plants in the edges of heavily wooded borders (less preferred by the beetles) maintain the general distribution of the beetles in an area after the weeds cease to be a range problem. That fact may assure the return of the beetles in effective numbers quickly enough to take care of reinfestations before they can reach serious proportions.

THE CONTROL OF A WEED by the biological method involves several aspects of ecology. Klamath weed is primarily a pest in range areas where soil moisture is ample from winter to early summer but deficient later in the year. Overgrazing fosters its spread. Under such conditions, its deep root system enables it to overcome even the sturdiest grass competitors, particularly when grazing has been so heavy that seed production by the more vigorous perennial grasses is curtailed.

An insect may control weeds by more subtle means than direct destruction. If its action is such as to remove the Competitive advantage of the weed host over desirable plant species, the weed may then be overcome by plants that cannot alone compete with it. That does not explain the control of Klamath weed, but the pressure of the beetles on the weed at a time when vigorous competing range plants occupy the area may be enough to prevent the return of the weed in such fields under proper grazing management.

Several investigators believe that three-fourths of the land south of Mount Shasta and from the coast to the Sierra foothills in California was originally covered with perennial bunchgrasses. Annual plants now make up most of the forage there. Their replacements by perennials would be impossible (and not necessarily desirable) everywhere in the region.

It therefore seems probable that with the destruction of Klamath weed the predominant annual-plant cover characteristic of the region may regain the land under normal conditions. That has happened in the areas where the weed has been cleared for three successive seasons. In Placer County, annual grasses, dominated by soft chess (Bromus hordeaceus), legumes such as birdsfoot trefoil, clovers, and lupines, and desirable fortis such as filaree, have returned as thickly as they are in neighboring range lands that have remained free of Klamath weed.

The success of the beetles in Humboldt County is attended by circumstances favorable to return of a forage cover of maximum value. In the areas most heavily infested with the weed (indicative of a favorable soil and site), the main perennial bunchgrass of earlier years (Danthonia californica) has managed to survive along animal trails and about the edges of seepage areas, which were too wet for Klamath weed in winter. The destruction of the dense stands of the aggressive weed has permitted a gradual return of this fine range plant. Although its distribution in the original beetle-release remained spotty for some years, it soon began to develop a vigorous cover that spread slowly over new ground.

At the end of a 50-acre field that had been cleared of weed by the beetles for 3 years, a rather complete stand of the hardy bunchgrass developed. Over the whole field most of which had been weed-free only 2 years Danthonia increased from 9.2 percent of the total plant cover in 1947 to 23.4 percent in 195o. By considering soft chess (Bromus hordeaceus), Danthonia, and desirable legumes together, one gets a general picture of the total forage improvement due to the beetle action. The three desirable types increased from 14.8 percent in 1947 to 43.4 percent in 1950. Klamath weed was reduced from predominance (57.6 percent) to complete absence. Thus the position of the weed and the position of the three desirable plants was practically reversed.

The Chrysolina beetles have become a permanent part of the natural fauna. Their future success will depend mostly on how closely the life processes of the beetles and their host coincide with changes in weather, for on that synchronization depends how fast the beetles multiply and how intensive is their action on uninvaded weed stands and reinfested fields.

Indications are that the beetles can duplicate throughout the Northwest the success they have had in California. It would be an economical, self-perpetuating way to combat a serious pest, now that the first intensive research and exhaustive explorations are completed.

JAMES K. HOLLOWAY is an entomologist in the Bureau of Entomology and Plant Quarantine, the division of foreign parasite introduction, and specialist in biological control in the University of California College of Agriculture. He has been engaged in research in biological control since 1927 and has had responsibility for carrying forward the project on the biological control of Klamath weed in this country since its inception in 1944. He studied at Mississippi State College and Ohio State University.

C. B. HUFFAKER, an entomologist and ecologist, has been engaged in ecological research since 1940. In 1946 he was appointed assistant entomologist in the division of biological control of the University of California, where he has been particularly interested in the population complexes of insects and the weeds attacked by them as components of a natural range environment. Dr. Huffaker holds degrees from the University of Tennessee and Ohio State University.

Mexican bean beetle larva.