Progress in Insect Classification

C. F. W. Muesebeck.

The accurate identification of an insect is the key to all past recorded experience with that species. Without it, costly mistakes may be made in the application of control measures, ineffective or unjust quarantine practices may be instituted, or much work may be unnecessarily duplicated. If there were only a few hundred, or even a few thousand, different kinds of insects it would not be very difficult for an entomologist to learn to recognize them all and to call their names but nearly 700,000 different kinds have been described and named, and it is estimated that at least twice that number remain to be identified.

Obviously it is quite hopeless therefore to determine what a given insect really is without the help of some orderly arrangement or classification of all the known kinds. To be sure, a comparatively small number of common and distinctive insects will always be readily recognizable without special aids, but the vast majority can only be identified by the skillful use of keys, descriptions, and other guides that result from the painstaking research of many specialists. It is this research in classification that makes definite identification possible. How accurate and complete the identification will be depends on how thorough and critical the research has been.

Classification of living things is an effort to interpret nature. It attempts to bring together the kinds that are alike and closely related and to separate those that are unlike and unrelated. The earliest classifications of insects were based largely on habits and habitats and on certain gross anatomical features that contribute to define the fades, or general aspects, of the different kinds. They were trial classifications and were naturally extremely artificial. They brought together things that were in no sense related and separated widely forms that belonged close together. With the growth of knowledge about insects and the rapid increase in the number of known kinds, however, the search for new characters usable in the development of more satisfactory classifications was intensified. Methods and concepts are improving steadily; the result is that insect taxonomists, gradually are producing in their classifications an interpretation of insect life that is much closer to existing facts than any of the arrangements previously developed.

In the efforts to attain the goal of having classifications in accord with the natural relationships of insects, one has to take into account not only anatomical characteristics but also facts pertaining to the physiology, biology, distribution, ecology, and sometimes cytology, of the species. Sometimes, indeed, anatomical distinctions are lacking or at least are not evident, although conspicuous differences have been observed in the life habits of the insect populations in question differences in time of appearance, food preferences, method of hibernation, or even in reaction to certain insecticides.

Such facts, when known, are usually indicators of fundamental distinctions between forms that at first appeared to be identical. They suggest that a restudy of series of specimens may reveal structural differences formerly overlooked, and often it does. That was so with the screw-worm flies, the European spruce sawflies, and the California red scale and the yellow scale of citrus, among others. The two screw-worm flies were long regarded as a single species that fed sometimes as a true parasite on warm-blooded animals, including man, and sometimes as a scavenger upon dead animals. These differences in feeding habits led taxonomists to a comparative study of flies reared from larvae of the two habit types, and to the discovery that the flies from the two sources could, after all, be distinguished on the basis of anatomical differences and represented two entirely different species.

That information resulted in an abrupt change in control procedure against the parasitic form. With the spruce sawflies it was a cytological study that established the distinctness of two species long regarded as one. In the case of the red and yellow scales of citrus, differences in susceptibility to attack by certain parasites and differences in location on infested trees eventually led to the discovery of structural differences by which the two species could be identified.

By the utilization, then, of all available information, taxonomists are attempting to correlate behavior and other life characteristics of insects with anatomical features, since it has come to be realized that only in this way can sound, natural classifications be developed. In the formation of keys, which are the guides to identification and which reflect the judgment of taxonomists with respect to relationships, however, it is necessary to depend on the use of physical characteristics of the insects. Only those are always definitely determinable from the specimens themselves. Therefore the principal efforts of the research taxonomist are necessarily directed toward the search for physical peculiarities, however small, that seem likely to be relatively constant and more or less distinctive. That that is no simple task must be obvious from the enormous number of known kinds and the continued addition to that number of 10,000 or more new species annually. A key that will infallibly lead the user to the correct name for a given insect can rarely be constructed even for a comparatively small group of insects. Few characteristics are absolutely fixed, and the extent and direction of the variation are themselves extremely variable. Bracon hebetor, an abundant and widely distributed parasite of certain pests of stored products, ranges in color from completely yellow to wholly black.

On the other hand, all the moths of all the species in the genus Rupela are without exception entirely white. One individual may be five times as large as another fully matured specimen of the same species. In another species, perhaps a closely related one, the specimens may be of rather uniform size. Details of sculpture or of wing venation may be strikingly constant or may vary widely. It is important to determine the range of variability in each instance, but that requires large numbers of specimens and these are not always to be had. Seldom, therefore, can a character that is employed in a key be considered absolute, and no key can be regarded as more than a temporary guide to identification. It will inevitably require modification, or even complete recasting, as knowledge of the particular group involved increases; its usefulness will depend to no small extent on the aptitude, experience, and perhaps even intuition of the user.

Dimorphism among adult insects of the same species is a common phenomenon. It is often a cause of serious difficulty in the development of classifications. The winged sex forms in termites and ants, for example, bear little resemblance to the wingless workers of the same species; in the mutillid wasps, which are popularly and inaccurately known as velvet ants, the winged males are so unlike the wingless females that their identity can only be established by biological association. That is true also of the cankerworm moths, in which the female is a grublike egg sac but the male is a normal winged moth. Such striking caste or sexual dimorphism occurs in various sections of all the major insect groups. Furthermore, since the biological association of conspicuously different castes or sexes of the same species may be difficult and slow, it sometimes happens that the male and female of a single species are long treated as two distinct things and are known under different names. Only field observations or biological studies can establish the facts in such cases.

More often than not it is in the larval stage that an insect is destructive. Because it is harmless and is seldom seen, the adult may be unknown to the grower whose crop is being damaged. In order that the right control measures may be applied, the insect has to be identified in its larval stage. For practical reasons, then, it has become essential, in the case of various insect groups, to supplement classifications founded on adults with keys to the larvae. The development of such keys is particularly difficult and slow because the identity of the larvae must first be definitely established. Otherwise, however good the key might be, it would not lead the user to the available information on habits and control of the pest. The name originally proposed for the adult insect is the clue, and the larva must be identifiable by the same name. For definite association of the larva with the adult, however, field or laboratory studies must usually be conducted, and these often demand facilities not readily available. Accordingly, the number of different kinds of insects for which this type of association has been worked out is very small and grows slowly. Even in the Lepidoptera (moths and butterflies) and the Coleoptera (beetles), where more work on immature forms has been done than in the other major groups of insects, fewer than 3 percent of the described species are known in the larval stage. Since about 1910, however, a great deal of progress has been made in this field of taxonomic work, and because emphasis has naturally been placed on the injurious species, most of the major pests are now identifiable in the immature stages.

Although, as I have indicated, the normal course is to base names on the adults and to develop the original and principal classifications from adult characteristics, there is one conspicuous exception. That involves the important group of plant pests comprising the family Aleyrodidae, the members of which are known as whiteflies. They are tiny insects that are not collected abundantly in the adult stage but are commonly seen fixed on leaves of infested plants when they are in the pupal stage. The family, which contains such devastating pests as the citrus blackfly, was long ignored, and most of the present knowledge concerning its classification has been accumulated during the past 5o years, the first significant and basic work being done by American taxonomists. Good characters upon which to base a classification were discovered in the pupae, and nearly the whole classification of the whiteflies is founded on this stage. In fact, it is rarely possible to identify adults in this group because adults have been definitely associated with the immature forms in only a few instances.

In keeping with its growing complexity, taxonomy has gradually become increasingly specialized, until now a worker usually confines himself to a single limited field, as, for example, aphids, or fleas, ants, biting lice, cutworm moths, leafhoppers, scale insects, termites, thrips, weevils, certain sections of the wasps or bees, grasshoppers, gall flies, or mosquitoes.