The Gall Wasp Genus Cynips: A Study in the Origin of Species/Physiologic Species

by Alfred Charles Kinsey

 

PHYSIOLOGIC SPECIES

The mutations that will always attract first attention involve such morphologic structures as we have considered in the previous section. On the other hand, there are three groups of physiologic characters in the genus Cynips which allow additional insight into the nature of species. These physiologic qualities are to be observed in the form of the gall, the life histories, and the host relations of these gall wasps.

The galls produced by the Cynipidae are, of course, plant tissue, but their form and structure depend largely upon the nature of the gall-producing stimulus which the insect puts into the plant. The precise source and nature of this stimulus is not known, but there seems every reason for believing that it partakes of the nature of an enzyme or hormone produced by some particular insect structure. To a limited extent the form of the gall does depend upon the plant tissue involved, for the possible transformations, as we have already pointed out (Kinsey 1923:21), are more restricted in certain tissues; but beyond that the form of the gall is dependent upon the species of the insect rather than upon the species of the plant. There must be nearly as many different kinds of gall-producing enzymes as there are species of gall wasps. When a single species of gall maker attacks more than one species of oak, the form of the gall is essentially the same on all of the hosts. This point has been so often established (e.g. Cook 1902, Kinsey 1920:365) that it is hardly necessary to accumulate more evidence. Nevertheless, the several cases in Cynips may be cited.

Multiple Hosts of Cynips

C. maculosa tritior	on Q. dumosa, Q. durata
C. echinus douglasii	on Q. lobata (and Q. Douglasii?)
C. echinus dumosae	on Q. dumosa, Q. turbinella
C. echinus schulthessae	on Q. dumosa, Q. durata
C. teres hildebrandae	on Q. dumosa, Q. durata
C. plumbea	on Q. arizonica, Q. oblongifolia, Q. Toumeyi
C. fulvicollis vorisi	on Q. macrocarpa, Q. bicolor
C. fulvicollis major	on Q. alba, Q. macrocarpa, Q. Michauxii
C. fulvicollis gigas	on Q. lyrata, Q. Mühlenbergii, Q. Michauxii
C. fulvicollis fulvicollis	on Q. alba, Q. Michauxii
C. dugèsi simulatrix	on Q. grisea, Q. undulata, Q. Gambelii, Q. arizonica, Q. oblongifolia
C. dugèsi brevipennata	on Q. grisea, Q. Gambelii, Q. fendleri
C. bella bella	on Q. grisea, Q. undulata, Q. pungens, Q. arizonica, Q. reticulata, Q. oblongifolia
C. mellea anceps	on Q. alba, Q. stellata, Q. floridana, Q. Chapmanii, Q. breviloba
C. mellea bifurca	on Q. stellata, Q. floridana
C. mellea litigans	on Q. stellata, Q. floridana
C. mellea Carolina	on Q. alba, Q. stellata
C. nubila nubila	on Q. arizonica, Q. oblongifolia
C. nubila russa	on Q. arizonica, Q. oblongifolia
C. nubila incompta	on Q. reticulata, Q. glaucophylla
C. villosa acraspiformis	on Q. grisea, Q. undulata, Q. arizonica, Q. Tourneyi
C. villosa expositor	on Q. arizonica, Q. grisea
C. villosa apache	on Q. arizonica, Q. grisea
C. gemmula fuscata	on Q. Michauxii, Q. Mühlenbergii
C. gemmula prinoides	on Q. prinoides, Q. Mühlenbergii
C. hirta hirta	on Q. Prinus, Q. Michauxii

It will be observed that in several cases a single species of Cynips occurs on such distinct oaks as Quercus alba, Q. macrocarpa, and Q. Michauxii; Q. alba and Q. stellata; etc. Such cases, commonly reported in literature on galls, are usually based on misdeterminations, but the present records have been carefully checked and may be verified from insect material in our collections.

 

Concerning the relative importance of insect and gall characters in the differentiation of species among Cynips:

 

52 species have insect structures more distinctive than galls,

24 species have galls more distinctive than insect structures,

17 species have galls and insects equally distinctive.

 

In more than half the genus the insects are more diagnostic than the galls, and there are even cases of distinct insects with practically identical galls. Most of these cases involve such closely related insects as we have called varieties in the systematic portion of this paper, but the following cases involve the second taxonomic category:

 

Cynips fulvicollis and C. plumbea

C. dugèsi and C. bella

C. pezomachoides and C. gemmula (certain var. only)

C. pezomachoides and C. hirta (certain var. only)

C. arida and C. mellea

 

More striking, however, are the species in this genus which are hardly distinguishable except by gall characters. It seems valid to recognize taxonomic groups based on such hereditary, physiologic characters if they involve large populations with distinct host relations and distinct geographic ranges. Nevertheless, current taxonomic studies are so often established upon morphologic bases that they would not recognize such species as the 24 physiologic Cynips listed above. In consequence we have seen much confusion of biologic data and serious invalidation of important conclusions.

Some of these physiologic species of Cynips deserve especial mention.

 

1. Cynips maculosa and C. mirabilis are hardly distinct insects, altho their galls (figs. 207-210) are very distinct. These two stocks have been separated since early in the history of the subgenus, but it is certain that most of their evolution has involved physiologic capacities of the insects.

2. Insects of Cynips echinus echinus and C. echinus douglasii seem absolutely indistinguishable. That the galls are very distinct is evident from figures 154-158.

3. Insects of Cynips echinus schulthessae and C. echinus vicina are very similar. A comparison of figures 151-153 and 156-158 will show how characteristic the galls of such species may be.

4. All of the European Cynips would certainly be considered three species on the basis of either the bisexual insects, the bisexual galls, or the agamic insects. The three would be Cynips folii-longiventris, C. divisa-disticha-cornifex, and C. agama, with the last not entirely distinct. I am convinced that without the distinctive galls (figs. 125-137) of the agamic forms of these insects they would never have been recognized as the six specific stocks which they really represent.

5. The insects of Cynips nubila appear identical with those of C. russa. The two galls (figs. 299-300) are alike in all respects except color. Nubila galls are wine-purple; those of russa are yellowish-russet. The first occurs in Arizona south of Tucson, the other north of Tucson (fig. 58). They occur on the same hosts (Q. arizonica and Q. oblongifolia), and occur within a few miles of each other at places in their ranges. The data do not warrant the admission of seasonal, climatic, or host factors as explanation of the differences in gall color. The two represent extensive populations existing in adjacent but distinct geographic areas, and must be considered as species. One of these species must have originated from the other by mutation as abrupt as that which gave rise to the short-winged species treated in the preceding section of this paper.

6. Cynips multipunctata conspicua and C. multipunctata heldae are hardly distinct insects, altho their galls are strikingly unlike in external form (see figs. 203 and 205-206). The close relations of the two are attested by their identical hosts and adjacent ranges (fig. 31), as well as by their morphologic identities, and heldae seems to be a species derived by physiologic mutation from conspicua stock.

The phylogenetically ancient standing of the gall characters of these cynipids is evidenced in the internal structures of the deformations. The agamic galls of all of the species of the genus are produced on the veins of the leaves of white oak, with the single exception of Cynips heldae, which occurs either on leaf veins, petioles, or young stems of the oak. In most instances the galls appear on the under sides of the leaves. Beyerinck's figures of Cynips folii (re-drawn in our figs. 113-117) show the order of transformation of the normal fibrovascular tissue, and indicate something as to the plant elements involved. Other European students of gall histology have included species of Cynips in their investigations. There is the work of Lacaze-Duthiers (1850-1853), Fockeu (1889), Hieronymous (1890), Küster (1900, 1911) and Weidel (1911). Following the suggestion of Lacaze-Duthiers, all these workers have found four fundamental zones of tissue in most cynipid galls. These zones have been called the nutritive, protective, parenchyma, and epidermal layers, and in the degree and character of the development of each of these the European students have seen an essential uniformity of structure among the European species of Cynips.

Unfortunately, most of these histologic studies were made on three common European Cynips: folii, longiventris, and divisa. Only Weidel (1911) has given us a discerning study of Cynips disticha (fig. 123), and there he recognized five layers of tissue instead of the traditional four. My own studies of the gross anatomy of the galls of the entire genus Cynips, summarized in figures 117 to 124, lead me to believe that Weidel's five layers are the correct basis of homology in this genus.

If it is remembered that Beyerinck's studies (figs. 113-117) show that these leaf galls originate in the phloem of the fibrovascular bundle, from which they develop outwardly usually thru the lower epidermis of the leaf, the following interpretations will seem warranted:

1. Nutritive layer. The innermost tissue of the gall, lining the larval cell. A distinct layer in young galls of many species, soon becoming reduced by the feeding of the larval insect (and probably by sorption by the other plant tissues) to a thin, broken layer of shrunken, partially empty cells. Poorly developed in any but the very youngest galls of Acraspis. Possibly directly descended from phloem.

2. Protective layer. A sclerified tissue that is best developed in the European sub-genus of Cynips. The cell walls are thickened, and the cells may contain crystalline materials. The larval cell wall of most cynipid galls is largely made up of protective layer to which the remnants of the nutritive layer are attached on the inside and some spongy parenchyma tissue on the outside. The protective layer may be a direct development from sclerenchyma tissue in the vein. Apparently absent in Acraspis.

3. Spongy parenchyma. Occupying the central portion and constituting the major material in all the spongy and more hollow oak apple galls of this genus. Poorly developed in the subgenus Antron and absent, as far as I can see, from the galls of the subgenus Acraspis.

4. Collenchyma. Lying directly beneath the epidermis. A second layer in which the cells have thickened walls and usually crystalline contents. The layer appears hard and compact-crystalline to the naked eye. Practically absent (by an unfortunate coincidence) from the three species on which the first European studies were based, but present in most other species of that subgenus and in the other subgenera of Cynips. Constituting the bulk of the material in the galls of the subgenus Acraspis, and well-developed as the compact outer layer of Antron. Cook (1904) and Cosens (1914) treated this layer in certain species of Acraspis as modified parenchyma, but this seems to be an attempt to maintain the four layers of the European workers. This collenchyma layer in the gall may be developed from collenchyma to be found in a similar position in the normal leaf.

5. Epidermal layer. The outer covering of the gall, including the fairly normal epidermis and all of the abnormal developments from it. Largely naked or at the most with stellate hairs in most groups of Cynips. With a peculiarly faceted surface in many species of Acraspis, in some cases with each facet terminated by a unicellular process which may be spiny or long and wool-like. Obviously a modification of the normal leaf epidermis.

The precise homologies of these tissues must be made by some botanist using modern technic. It will be interesting to compare structures in galls of some of the species which occur either on upper and under surfaces of leaves, on veins, petioles, and (as in heldae) on young stems.

The distribution of the five types of gall tissues among the species of Cynips may be summarized as follows:

Gall structures in Agamic Cynips

0 = absent, — = poorly developed, + = distinct, + + = well developed.
Subgenus	Specific Stock	Nutritive layer	Protective layer	Parenchyma layer	Collenchyma layer	Epidermal layer
Cynips	folii	+	+	+ +	0	+
	longiventris	+	+ +	+ +	—	+
	divisa	+	+ +	+ +	0	+
	agama	+	+	+ +	0	+
	disticha	+	+ +	+ +	+	+
	cornifex	+	+	+	+ +	+
Antron	echinus	+	+	—	+ +	+
	guadaloupensis	+	+	+ +	+	+
	teres	+	+	+	+ +	+
Besbicus	multipunctata	+	+	+	+	+
	maculosa	+	+	+	+	+
	mirabilis	+	+	+ +	+	+
Philonix	plumbea	+	+	+ +	—	+
	fulvicollis	+	+	+ +	—	+
Atrusca	dugèsi	+	+	+ +	+	+
	bella	+	+	+ +	+	+
	cava	+	+	+ +	+	+
	centricola	+	+	+ +	+	+
Acraspis	arida	—	+	+	+	+
	mellea	—	+	+	—	+
	conica	—	+	+	+	+
	nubila	—	0	0	+ +	+ +
	villosa	—	0	0	+ +	+ +
	gemmula	—	0	0	+ +	+ +
	pezomachoides	—	0	0	+ +	+ +
	hirta	—	0	0	+ +	+ +

Within each subgenus there is striking uniformity in the degree of development of each gall tissue. Thus, in the subgenus Cynips the protective layer is unusually thick and the collenchyma layer is absent or poorly developed (except in the unique C. cornifex). In Antron all five layers are well developed, with a more pronounced development of the collenchyma layer. In Besbicus all five layers are again present (with the parenchyma most so in mirabilis). In Philonix the development is chiefly that of a rather solid spongy parenchyma. In Atrusca it is a parenchyma with few but tremendously extended fibers. In Acraspis all five layers are developed among the species centering about Cynips mellea, but in the other specific stocks in Acraspis the protective and parenchyma zones seem absent and the collenchyma and epidermal layers are unusually well developed.

The gall-producing stimulus, whatever its origin, is evidently selective in its effects upon particular plant tissues; and, since the gall characters are usually of subgeneric significance, it is apparent that these peculiarities of the gall-producing stimuli are of as ancient standing as any of the morphologic structures of the insects. This means that these physiologic qualities have been constant in heredity for possibly ten or twenty million years during which the specific stocks of Cynips have been differentiated.

 

The second body of data on the physiologic inheritance of species is concerned with life histories in Cynips. Like most of the other higher gall wasps, these insects have an alternation of a bisexual and an agamic generation. Thruout the groups the life history data are so uniform that it may be readily summarized.

The agamic insect develops in a gall which appears on the leaves early in the summer. The gall is mature by the end of the summer, and the insect matures and transforms into an adult early in the fall. Then—the most unique feature of the genus—the adult continues in the gall, chewing an exit passage thru everything except the epidermis of the structure, but not emerging until very late in the fall or some time in the winter. Most of the emergence is in mid-winter. The agamic females oviposit within the scales of unopened buds on the trees. The bisexual galls do not begin development until the leaves begin to unfold on the trees in the spring. These galls are simple, seed-like or bladdery, thin-shelled developments usually not larger than the buds within which they occur. The bisexual insects mature within three or four weeks, emerging in the middle of the spring. They copulate and oviposit in the main veins, usually on the under surfaces of the unfolded but still immature leaves which are on the oaks at that season.

The constancy of this life history thruout nearly all of the genus Cynips is a remarkable testimonial to the hereditary stability of physiologic characters. This precise combination of characters is found nowhere else among the gall wasps. Its nearest approach is in the genus Disholcaspis, which is certainly a close relative of Cynips. But the agamic Disholcaspis emerges in the late fall, ovipositing in the veins of the embryonic leaves instead of in the bud scales; the bisexual Disholcaspis females do not emerge until mid-summer, and they oviposit in the bark of younger twigs of oak.

The life history of any cynipid is determined by a variety of physiologic characteristics of the insect. The date at which the adult matures depends both upon the season at which it started development and upon its rate of development. The place of oviposition of the female depends upon the inherent reflex and tropistic responses of the insect to factors which lead them to particular parts of particular species of plants and which inspire oviposition in those places. The length of the dormant period thru which each egg goes before it finally hatches must depend on the nature of the egg materials and on the reactions of those materials to external factors. In Cynips the eggs go thru a mid-winter dormancy of say 16 weeks before they hatch. The eggs of Disholcaspis require about 22 weeks. In each group the date of hatching is, however, a matter of long standing.

The mid-winter emergence of the agamic Cynips deserves further consideration. It is an interesting fact that we have bred most of our 17,000 insects of this genus out-of-doors at temperatures never more than perhaps fifteen degrees above freezing, and in many cases ten or fifteen degrees (Fahrenheit) below freezing. When these agamic forms are at room temperature they emerge later than normally, if they emerge at all. Active insects brought indoors become more active for a few minutes, but soon they are killed by such stimulated activity.

The cause of mid-winter emergence is not satisfactorily explained, altho it has been generally accepted as inherent in some way within the species. Miss Payne's studies (1925-1926) on the behavior of insect tissues at low temperatures explain some of our phenomena, but there is much need for precise measurements and bio-chemical investigations on our particular winter-active species. Concerning European Cynips Kieffer (1901:632) devoutly remarked of the insect which had delayed its emergence: “Ce qu'il attend, c'est l'époque que la Nature lui a assignée.”

Whether the emergence date is determined by the response of the insect to environmental factors should be determinable by modifications of those factors. This experiment we have conducted on an extended scale, incidental to the breeding of Cynips material. We have brought 60 of the species of this genus from every type of remote locality to breed out-of-doors under the peculiar conditions of southern Indiana winters. We have bred material from northern Michigan and the mountains of northern California, from southern Georgia and the Gulf Coast, from southern California, Denmark, and more southern Europe. In many cases we have bred the same species for several winters, and in the case of Cynips fulvicollis (detailed later) we have had to keep transplanted material for two and three years outside our laboratory windows before the insects matured. Nevertheless, in all of this work, we have secured emergence at dates that would have been normal in the native habitats of the species. The season of emergence has been shown to be a specific quality which is not dependent upon the responses of the individual insects to immediate environmental conditions.

 

Northern Michigan material of Cynips pezomachoides wheeleri, due to emerge in northern Michigan during cold weather in late November and early December, emerged in southern Indiana at those same dates, altho at that season the temperatures are still very mild in our part of the country. Material of Cynips pezomachoides pezomachoides from near Boston, Massachusetts, emerged at Boston late in November and early in December at temperatures ten or fifteen degrees below freezing, while our material of apparently the same insect from the Carolinas, Georgia, and northern Florida emerged at the very same dates during mild weather in southern Indiana. We have had similar experience with several varieties of Cynips fulvicollis and still other species of the genus. The data for fulvicollis are:

Emergence of Cynips Fulvicollis

Variety	Source of Material	Chief Emergence Dates at Bloomington, Ind.
canadensis	No. Mich.	Late Nov., early Dec.
fulvicollis	So. Mich., no. Ind.	Early to late Dec.
	So. Ind., so. Ill.	Mid- to late Dec.
major	So. Ind., so. Ill., so. Mo.	Late Dec., early Jan.
vorisi	So. Kans.	Early Dec., to late Feb.

The most interesting aspect of fulvicollis is its earlier emergence in northern varieties and its later emergence in southern varieties. If, as our data show, the emergence date is not determined by direct response of the individual insects to environmental factors, then how may we explain this apparent correlation of emergence dates with the latitudinal range of each species? If Lamarckian effects enter here, they must come as the result of influences continued over numerous generations. Is it more probable that Darwinian selection has helped adjust these species to climatic conditions? Or is it possible that the apparent correlation is fortuitous?

All but two of the groups of species of Cynips complete their agamic lives within six to eight months. In Cynips mellea emergence is spread from the seventh to the tenth month (December thru March) with most of it in March. This is a departure from the ancestral tradition, but one specific for mellea. Cynips fulvicollis is a further departure, for most of its emergence is delayed until the eighteenth to twentieth month. This is particularly true of the more northern varieties of the species. The Kansas variety, vorisi, which is nearest the point of origin of the group, has the most normal life history, most of the individuals emerging during the first winter with only a few of them delaying emergence until the second season. The northern variety canadensis gives no emergence in the first year, most of it in the second year, and some stray emergence in the third year. The varieties ranging between vorisi and canadensis show a gradual shift from the one to the three-year cycle. It is an interesting case of new specific characters developing out of physiologic materials that have remained constant in other groups of Cynips for millions of years.

The third body of physiologic data bearing on the nature of species is to be drawn from the host relations of our present insects. Every species of the 93 in the genus occurs on white oaks, including, however, oaks of both the groups Leucobalanus and Protobalanus of the Trelease classification (1924). In common with many other Cynipidae, these insects thus throw doubt on the validity of separating Protobalanus from the other white oaks.

The restriction of our genus, as here classified, to the white oaks is of especial interest because previous monographs of the group (Dalla Torre 1893, Dalla Torre and Kieffer 1910, and Beutenmüller 1911) have included many black oak species. Here is an illustration of the importance of basing biologic conclusions on sound taxonomic classifications.

The most significant of the host relations of long phylogenetic standing within the genus may be summarized as follows:

 

1. The Pacific Coast subgenus Beabicus represents three stocks, multipunctata, maculosa, and mirabilis, which are restricted to the groups of oaks centering about Q. lobata, Q. dumosa, and Q. garryana respectively. The subgeneric stock must have reached the Sierras, where it split into the three stocks, before the Great Basin became arid during the Miocene. Today, every individual of all the species of this subgenus shows sensory reactions of essentially the same nature as those shown by the ancestral stocks many millions of years ago.

2. Similarly, the ancestral stocks of Cynips echinus and C. guadaloupensis, of the Pacific Coast, have perpetuated their host preferences, Q. lobata-dumosa and Q. chrysolepis respectively, thruout the 9 species which these groups now represent.

3. One of the most special host restrictions in this genus is that of the four species of the Cynips centricola group on Q. stellata. This evidences the persistence in heredity of a specialized physiologic character.

4. In the Cynips mellea group, 8 of the 11 species similarly occur on Q. stellata. In this case, however, one may find stray individuals on Q. alba and other hosts, especially in regions where Q. stellata is rare or lacking.

5. The species grouped under Cynips pezomachoides and C. gemmula are obviously very close relatives. Nevertheless, the seven species of pezomachoides east of the Great Plains are confined to Q. alba and its close relatives while the three species of gemmula occur on chestnut oaks of the Q. Prinus group. The ancestors of each group must have been separated on the basis of host preferences, and their descendents still maintain the ancestral choice.

 

It would appear, then, that species in nature may be differentiated wholly or largely either on physiologic or morphologic bases, or equally on both bases; and it seems possible to recognize species of Cynips that have originated by physiologic as well as morphologic mutation. This seems reasonable, for organisms inherit their psychologic and physiologic characters in the same sense as they inherit their morphologic structures. As a matter of fact, the materials transmitted from one generation to the next are neither morphologic nor physiologic characters but an initial bit of simple protoplasm and a physico-chemical organization which will direct the activities of that protoplasm. Many outside, environmental factors also affect the developing organism, but the final form of the plant or animal is inherited in the sense that the inherited genes exert the primary influence on that form. It is in precisely the same sense that the physiologic or psychologic characters of species may be said to be inherited. There seems no sound basis for the oft-made suggestion that physiology is a function of structure—or structure of physiology. It would appear that both are products of the same protoplasm and are controlled by the same hereditary mechanism.