%r QH371 L8 (!;fa> B. U. Ml library Norttj (Earoltna &tat? Intweraitij 371 L3 THIS BOOK IS DUE ON THE DATE INDICATED BELOW AND IS SUB- JECT TO AN OVERDUE FINE AS POSTED AT THE CIRCULATION DESK. 0. H. I H. C. ST 100M/5-79 EXPERIMENTS WITH DROSOPHILA AMPELOPHILA CONCERNING EVOLUTION. BY FRANK E. LUTZ PUBLISHE :* BY TH WASHINGTON, u. c. e Carnegie Institution of Washington 1911 North Carolina State Library Gift of North EXPERIMENTS WITH DROSOPHILA AMPELOPHILA CONCERNING EVOLUTION. BY FRANK E. LUTZ WASHINGTON, D. C. Published by the Carnegie Institution of Washington 1911 CARNEGIE INSTITUTION OF WASHINGTON, PUBLICATION No. 143 Paper No. 16 of the Station for Experimental Evolution at Cold Spring Harbor, New York Copies of this were- first issued MAR 14 1911 the cornman printing CO. CARLISLE, PA. EXPERIMENTS WITH DROSOPHILA AMPELOPHILA. THE INHERITANCE OF ABNORMAL VENATION. THE EFFECT OF SEXUAL SELECTION. DISUSE AND DEGENERATION. in THE INHERITANCE OF ABNORMAL VENATION. Practically all the experimental studies of inheritance have extended through but few, rarely more than 6, generations and have been con- cerned with pairs of non-intergrading characters. In the present work more than 70 generations have been reared. This was possible for two reasons: Drosophila ampelophila Loew has a very short life-history, and it can be kept breeding throughout the year. The character abnormal wing-venation, the inheritance of which was studied, may be made to exhibit extreme variability, passing from less venation than normal through normal to extra venation, so great that the additional veins almost equal the normal in extent. At the Boston (1907) meeting of the International Zoological Congress a preliminary report was presented upon this subject, 6 generations having been obtained. During the summer of 1908 a report upon the work (covering about 25 generations) done at the Station for Experi- mental Evolution was submitted to the Director, but I deferred publi- cation because I wished to test more in detail certain points, especially sexual selection and the further fate of the abnormal strains. This additional work was done at the American Museum of Natural History. Incidentally I obtained confirmation of the previous work, but for the most part the present paper includes only the Cold Spring Harbor data and the conclusions drawn are as given in the 1908 report, except where otherwise indicated. MATERIAL AND METHODS. Drosophila ampelophila (the small red-eyed "pomace-fly") is very common about cider-mills, ripe fruit, vinegar-barrels, and the like. The larvae normally live in the pulp of rotting fruits, especially during the acetic-acid stage of decay. They will, however, thrive on the side of a tumbler containing fruit-juices, and I have reared them through several generations on stale beer. At a temperature of 25° C. the eggs hatch in 40 hours or less. The duration of the larval period is, on the average, 5 days, 2nd of the pupal period 4§ days. The adults become sexually mature about 48 hours after emergence when kept at this tem- perature. They live for about 3 weeks. The mean number of eggs is close to 200. Copulation is repeated and frequent. Most of the flies discussed in this paper were bred in an incubator, where an average temperature of 25.5° C. was maintained. A thermo- graphic record was kept. Since the temperature of the incubator was so nearly that of the working-room, absolute constancy was not obtained. The amount of variation is shown in fig. 1, which gives the frequencies l 2 EXPERIMENTS WITH DROSOPHILA AMPELOPHILA. of the different degrees as found during four typical months from read- ings of the thermogram at 3-hour intervals. For the purpose of these experiments even this approximation to constancy does not seem neces- sary, as variations of temperature were found to have no influence upon the wing-venation. Therefore the incubator was not used in the latter part of the work. Class 18.33 Frequency o Fig. 1. Bananas were used as food. They were purchased while still quite green and ripened in glass-stoppered bottles. In this way accidental introduction of wild flies was rendered unlikely. Even had Drosophila eggs been laid on the green banana, they would have hatched and the larvae would have developed into plainly visible pupae before the banana was used. Frequent control-cultures were kept and in no case was a Drosophila found in them. The flies with their food were kept in care- fully washed glassware and the instruments used in handling the food were sterilized in an alcohol flame after every operation which could possibly get eggs or larvae upon them. The importance of this caution can not be too strongly urged upon those who carry out pedigree- work with this insect. An egg-laying female was given a fresh piece of banana every two days and an effort was made to have all the banana of the same degree of decay. Each piece was kept separate during the growth of the larvae. This also is important, since, if one merely gives a large supply of food to the female at the start of oviposition, and does not change it, the early-born larvae will have very different food from those which are born later. The pupae were picked out of the "larval dish" and placed INHERITANCE OF ABNORMAL VENATION. 3 upon moist blotting-paper in a small vial, from which the adults could readily be transferred to an etherizing vial as they emerged. When mating was to be done the sexes were always separated before they were a day old. Usually no female was used as a parent that was more than 12 hours old before being isolated from the males. Numer- ous tests showed that no females so treated laid fertile eggs. Only rarely was there a difference of more than one day in the ages of the parents, and they were usually mated before they were two days old. For practical reasons, parents were killed after 50 to 100 offspring had been secured. It was found that neither the percentage of abnormal offspring nor the intensity of their abnormalities changed with the age of their parents, so that this procedure was permissible. In this paper only those families are considered which are in or close to the main line of descent. I have not thought it worth while to include any families having less than 40 offspring unless they were in this main line. Typical data are given in table 36, page 31. I have tried to arrange these so that they will be available for further work by those interested. They should not, however, be used for more than they are worth. For example, one can not study the inheritance of fecundity from them, as in but few cases have I bred from a female until she died a natural death. All individuals, both parents and offspring, have been kept for refer- ence and are deposited in the American Museum of Natural History. When of especial interest, the wings were mounted on glass slides in a thin layer of paraffin. This was found to be an excellent method of preservation. By all other methods which were tried the veins were rendered more or less transparent. When, as in making matings, it was desired to examine live flies, they were slightly etherized. They completely revive in a few minutes. All examinations for abnormalities in wing- venation must be made with a lens. Occasionally the larvae were attacked by a disease (?) of unknown origin which caused them to crawl out of the food, elongate, and die. When this disorder appeared in a dish it was usually fatal to all the larvae in that dish. Otherwise, Drosophila bears confinement very well. Prac- tically all the larvae which hatch complete their development. My ex- perience confirms the results reached by Castle (19066) that the closest inbreeding may be practiced with this fly for generations with no injuri- ous results. Such inbreeding was the rule in this work, being necessary in long-continued breeding unless unpedigreed stock be used. DESCRIPTION OF NORMAL VENATION. The normal venation of Drosophila is extremely simple, as is shown by fig. 2. The costal vein reaches to the fourth of the five longitudinal veins. The auxiliary vein is incomplete or indistinct. The anal cell is present. The discal and second basal cells are united and the first pos- terior cell is not appreciably narrowed in the margin. 4 EXPERIMENTS WITH DROSOPHILA AMPELOPHILA. DESCRIPTION OF ABNORMAL VENATION. It is probable that all insects occasionally show some abnormality of wing-venation. In my experience with Drosophila ampelophila they occur in one-third of 1 per cent of wild specimens. The data concern- ing this point are given in table 1. In these the abnormalities consisted of irregularities of the second longitudinal vein or small dashes near its distal end (similar to figs. 3 to 10). Only one of the 19 abnormal* wild flies I have seen was abnormal in both wings. Table 1. — Percentage of wild Drosophila ampelophila which have extra veins in their wings. Normal. Abnor- mal. Percent- age of abnormal. Bloomsburg, Pa Huntington, N. Y. ... Woods Hole, Mass.... Boston, Mass 1165 697 2083 1660 8 3 3 5 0.68 0.43 0.14 0.30 Total 5605 19 0.34 While rearing this insect for another purpose, several such abnormal specimens were found in one family. My principal abnormal strain, in which the variety and amount of abnormality is little short of astound- ing, came from these. The various figures give a better conception of what was obtained than would verbal description. There is the utmost variation in the abnormal venation, not only in different flies, but in the different wings of the same fly. The majority of the abnormalities are in the distal portion of the marginal cell, but they have been found also in the submarginal and a few in the first, second, and third posterior cells, affecting all the longitudinal veins except the first. CORRELATION BETWEEN THE RIGHT AND THE LEFT WINGS. One wing may be abnormal, or both may be. In the latter case the abnormality may be great in one wing, small in the other ; on one vein in one wing and lacking on this vein but present on others in the other .wing (see figs. 44 to 46). Nevertheless there is a correlation between the intensity of the abnormality in the two wings, as is made clear by tables 2 and 3. In drawing up these tables the range of variation of the intensity of the abnormality was divided arbitrarily, since the char- acter is not quantitatively measurable, into six classes: normal vena- tion (or zero intensity of the abnormality), very slight (see figs. 3 to 7), slight (see figs. 8 to 11), medium (see figs. 12 to 18), great (see figs. 19 to 24), and very great (see figs. 25 to 35). To which class a given wing should be assigned is a matter of judgment ; but since when these tables were made up it was thought that there was no correlation be- *Unles3 otherwise stated "abnormal venation" means, throughout this paper "veins added." & F F ' INHERITANCE OF ABNORMAL VENATION. tween the two wings with respect to the intensity of abnormality, the personal equation which entered in would have tended to make the cor- relation as shown by the tables too low rather than too high. The arabic Table 2. — Correlation between right and left wings of males. [For explanation see page 4.] RIGHT. N. V. S. S. M. G. V.G. N. 413 340 48 79 33 46 34 58 3 17 531 V. S. 48 60 23 12 8 8 14 10 1 3 94 S. 23 46 15 9 9 6 15 8 9 2 71 M. 20 47 14 10 17 7 14 8 9 2 74 G. 3 13 3 3 2 2 9 2 2 2 1 20 V. G. 1 1 1 507 103 69 86 25 1 791 Table 3. — Correlation between right and left wings of females. [For explanation see page 4.] RIGHT. N. V. S. S. M. G. V.G. N. 226 150 43 51 37 57 47 73 1 15 5 354 V. S. 42 46 22 16 21 27 20 25 3 6 1 2 109 s. 25 54 20 18 33 21 37 26 12 7 1 2 128 M. 42 67 25 23 33 25 46 10 8 2 2 158 G. 9 21 7 7 6 8 13 20 11 5 4 1 50 V.G. 6 2 2 5 3 6 4 15 344 117 130 168 43 12 814 numbers show the observed conditions ; the italics show the distribution of frequencies which would have been expected had there been no cor- relation. The fact that expectation is exceeded by observation in those classes where the intensity is alike, or nearly so, in each wing, but is not EXPERIMENTS WITH DROSOPHILA AMPELOPHILA. INHERITANCE OF ABNORMAL VENATION. 46 a 8 EXPERIMENTS WITH DROSOPHILA AMPELOPHILA. equaled in the classes where the two wings markedly differ, indicates a definite positive correlation. Furthermore, when one wing is abnormal the chances that the other one will be abnormal also are 62 in 100 in the case of the males and 74 in 100 in the case of the females. This is an estimate based upon 4,000 pedigreed individuals. It will probably not hold for wild flies, since a large part of the 4,000 were from the abnormal strain; hence the esti- mated chances are larger than they would be in nature, because, as will be shown shortly, there is a close relation between the percentage of abnormal offspring in a family and the likelihood that an abnormal fly will be abnormal in both wings. It does, however, give an idea of the correlation which exists between the two wings with respect to the presence or absence of abnormal venation when such abnormalities are well fixed, and it brings out the further point that there is a sexual dif- ference to be considered. SEXUAL DIMORPHISM. The females show a greater tendency to be abnormal than do the males, and, when abnormal, their abnormalities are, on the average, more intense than those of the males. The first of these points is illus- trated in table 4 and fig. 51. Table 4 shows the percentage of abnor- mal males and females in 200 families. It will be noted that as the percentage of abnormal males increases the percentage of their sisters which are abnormal increases until the latter have become practically 100 per cent abnormal. Then, since they can go no further, their brothers gain on them in abnormality until we get families in which 100 per cent of both males and females are abnormal. In fig. 51 the crosses show the position of the mean percentage of abnormal sisters for each 10 per cent grade of abnormal brothers. A line is drawn to show the condition when for each per cent of male abnormality the female abnormality is 1.5 per cent. Thus, when 40 per cent of the males are abnormal, 60 per cent of their sisters are abnormal. Corresponding to 60 per cent male abnormality, we get 90 per cent female abnormality. Beyond that the females can go little further, hence the line becomes INHERITANCE OF ABNORMAL VENATION. TABLE 4. — Percentage of abnormal males and females in 200 families. Females .0 5.0 15.0 25.0 35.0 4S.O 55.0 65.0 75.0 85.0 95.0 .0 31 II 1 43 5.0 2 10 7 2 21 15.0 1 2 3 5 1 12 25.0 3 2 3 4 12 350 3 3 2 3 2 1 1 15 45.0 3 1 3 5 ' 12 55.0 A i 3 2 10 65.0 1 2 l 5 9 75.0 1 3 6 3 f3 8S.O 3 6 9 95.0 l 4-3 44 33 22 10 8 10 7 6 12 10 17 65 200 Females n n 2 0 5 15 25 35 45 55 65 75 85 95 o >%,■ + 5 15 N. + 25 s + 35 \ + 4-5 + 55 ■KS. 65 4 75 + 85 + 95 + Fig. 51.— For explanation see p 8. 10 EXPERIMENTS WITH DROSOPHILA AMPELOPHILA. vertical. The close fit of this line to the observed data shows that the relation Percentage of abnormal females = 1.5 X percentage of abnormal males may be taken as approximately describing the average observed con- dition. Owing to the impossibility of describing the intensity of abnormality in quantitative grades, we can not give a formula for showing its sexual relation. Tables 2 and 3 show that there is such a relation. The ques- tion as to whether both wings or only one shall be abnormal is also a part of this same problem of the variation of the intensity of the abnor- mality. We have seen that when a female is abnormal she will in 74 per cent of the cases be so abnormal that both wings will be affected, while only 62 per cent of her abnormal brothers will be abnormal in both wings. THE RANGE OF VARIATION OF ABNORMALITY INCLUDES "NORMAL" VENATION. One other point is to be noted. The intensity of abnormality ranges all the way from cases in which there is almost as much abnormal vena- tion as normal down to a barely discernible devia- Table 5 tion from normality. We have, then, in studying the inheritance of abnormal venation, the serious difficulty that a just indiscernible abnormality may be present.* Such a fly would be recorded as nor- mal. Table 5 suggests that they would be more like- ly to occur in families in which the percentage of abnormal offspring is low, for as such percentage decreases the percentage of abnormal individuals which are abnormal on both sides (C. S. ) decreases. In other words, there is an increasing percentage of abnormal flies which have the abnor- mality so reduced that in at least one wing it can not be seen. Hence, presumably, there is an increasing percentage of flies which have the abnormality reduced in both wings to a point just below visibility. These will be more common among males than among females, because the intensity of the abnormality is less in male than in female wings. Whether this alone accounts for the fact that a smaller percentage of brothers are visibly abnormal than of sisters is a question to which it is difficult to give an answer. • *M^l ^°i this be true aIso of the sP°tted condition in certain mammals? A guinea- pig still behaves as a spotted animal even if the spots are reduced until only the eyes remain affected. If the variation goes still further we would have an animal germi- " 1 t S^°"te somatlcally sPotless- We would then say that the spotted condition is | Percentage of • abnormal flies C. s. per family. lto 20... 20.9 21 to 40... 30.9 41 to 60... 50.6 61 to 80... 58.0 81 to 100... 83.1 INHERITANCE OF ABNORMAL VENATION. 11 HISTORY OF THE PEDIGREED STRAIN. Before taking up the data concerning inheritance, it will be well to out- line briefly the history of the chief pedigreed strains. Further details are given in table 36. Mating 211 was the first family in these lines of which a large number of offspring were described. Both parents were abnormal in both wings. The wings of 177 offspring of this mating were sketched. It was found that 31 per cent of the males were abnormal and 65 per cent of the females. Successive generations after this, breeding brother with sister, gave the following results: Abnormal female by normal male (mating 257) , 70 per cent of each sex abnor- mal; abnormal female by abnormal male (mating 284), 62 per cent of the males and 96 per cent of the females abnormal; abnormal female by normal male (mating 330), 96 per cent of the males and 91 per cent of the females abnormal; abnormal female by normal male (mat- ing 367) , 64 per cent of the males and 91 per cent of the females abnor- mal. A number of matings were made from the offspring of No. 367. Matings 405 and 408 are of especial interest. In both of these matings both parents were abnormal in both wings. Unfortunately there were a small number of offspring from each (25 and 29, respectively), but all of the offspring of mating 405 were normal and all those of mating 408 were abnormal. Three matings were made from the offspring of 405. Of the 385 offspring of these, not a single one showed the slightest trace of an abnormality, while of the 51 offspring of mating 440 (the parents being children of 408) only one, a male, was free from abnormal venation. Mating 405, then, became the starting-point of the "normal strain" and mating 408 the starting-point of the "abnormal strain." As can be seen from table 36, the various generations of the abnormal strain gave approximately, sometimes actually, 100 per cent abnormal flies, although normal individuals were far from rare. Furthermore, the intensity of the abnormalities increased. The greatest abnormality noticed before the fifth generation is shown in fig. 20. Up to that time all abnormalities were confined to the second longitudinal vein. Begin- ning with the sixth generation, abnormalities appeared on the third longitudinal vein. They became frequent by the tenth generation. In the fifteenth generation they were common and abnormalities began to be noticed on the fourth longitudinal vein. These have, even yet, rarely exceeded small spurs near the distal end. About this time the fifth longitudinal vein also began to be affected, and specimens such as are illustrated in figs. 30 and 32 were found. Meanwhile increasingly great abnormalities on the second and third longitudinal veins occurred. (See figs. 37 to 42 for examples. The condition shown in fig. 43 is unique. ) Turning now to the normal strain, three points should be borne in mind : the parents in each generation were normal, it came from the same 12 EXPERIMENTS WITH DROSOPHILA AMPELOPHILA. ancestry as the abnormal strain, and about one- third of 1 per cent of wild Drosophila ampelophila were found to be abnormal. For four generations after branching from the abnormal strain (five, counting mating 405) not a single abnormal individual was found, but in the next generation 1 fly out of 216 (0.5 per cent) had a very slight abnormality. In succeeding generations the percentage increased for a time, in spite of artificial selection to the contrary, and then diminished to zero under the same treatment. The abnormalities were all small, never greater than "medium." Table 6 summarizes the history of this strain for 40 generations. Table 6. — Fluctuation in percentage of abnormal individuals in a normal strain. Generations of normal strain. No. of normal. 1 and 2 3 4 5 7 9 11 13 15 17 19 21 6 8 10 12 14 16 18 20 22 418 235 575 594 336 471 183 79 110 234 56 No. of Percentage ! of abnormal. abnormal. 0 0.0 0 0.0 1 0.2 7 1.2 31 8.4 45 8.7 14 7.1 2 2.5 5 4.3 14 5.6 2 3.4 Generations of normal strain. 23 and 24 25 27 29 31 33 35 37 39 26 28 30 32 34 36 38 40 Total No of No. of Percentage of abnormal. normal. abnormal. 144 0 0.0 239 0 0.0 257 1 0.4 116 0 0.0 195 0 0.0 217 0 0.0 148 0 0.0 55 0 0.0 63 0 0.0 4,725 122 2.5 It is to be noted that the percentage of abnormal individuals is greater in this strain artificially selected for normal venation than it is in nature. At first thought one would say that this is an effect of the environment to which they were subjected. If this be true, environment may have played a part in the production of the abnormal strains. I think, how- ever, that it is not true. A sufficient explanation seems to lie in the fact that a form of selection exists in nature which is keener than the artificial sort, even when the latter is carried out under a lens. THE EEFECT OF SELECTION. Of late years there has arisen considerable skepticism concerning the cumulative effect of selection except as a means of isolating "pure lines." Jennings (1908) says : "Certainly, therefore, until some one can show that selection is effective within pure lines, it is only a state- ment of fact to say that all experimental evidence is against this." Whether or not the present material has a bearing upon the question thus clearly put depends upon the definition of a pure line. If a pure line be defined as one from which nothing else can be gotten by selec- tion, further discussion is not necessary. On the other hand, if inbreed- ing (for the most part, brother X sister) for 10 or 15 generations and rigid selection (in this case, with respect to wing-venation) may be INHERITANCE OF ABNORMAL VENATION. 13 reasonably supposed to have established as " pure " a line as exists in a given case, the following facts may be of interest. The abnormalities obtained, both in the direction of veins added and of veins lacking, far surpass those found in nature in this or any other insect with which I am familiar. Furthermore, they do not even remotely suggest the venation of any of this fly's relatives. Something new has been produced. In the strain whose early history has just been described there was, at the start, no definite effort made to build up an abnormal race as quickly as possible. Later I tried to do this from wild material obtained from other localities. Starting with an abnormal male and a normal female from Boston and an abnormal male and female from Bloomsburg, Pennsylvania, I rigidly selected for additional veins. The record for each successive set of two generations was 8.8, 5.5, 11.5, 14.3, 30.3, 45.8, 85.9, and 100 per cent abnormal. Thereafter mass-breeding was practiced and the abnormal strain preserved for about a year by merely starting a fresh jar every couple of weeks with the most abnormal individuals found at that time. The abnormalities in this strain were of the same nature and extent as in the one started from the Long Island material. It would seem that this increase in the percentage of abnormal individuals up to 100 per cent and the subsequent increase of the intensity of the abnormalities can not be due to the gradual weeding out of all units but the one or several desired, because one quickly gets things which one can safely say did not exist in the population with which we started, or, to be more exact, which we do not see. Some can probably imagine that the ' units" for each successive grade of abnormality existed in the parents with which we started, but that they were held in check by an equal number of inhibiting "units " of corresponding powers, so that the result could be explained by saying that in the selection we cut out step by step suc- cessively stronger inhibiting units, thus allowing successively greater abnormality-producing units to manifest themselves. On any other hypothesis, it seems to me, we must admit the cumulative effect of selection upon a "unit," i. e., within a pure line. But, upon this hypothesis, how can we account for the occasional nor- mal flies? Why do not the inhibiting units stay cut out after we have once gotten rid of them so thoroughly that all the flies of several suc- cessive generations show strong added veins? Perhaps they do stay cut out and these occasional normals are merely fluctuating variations in the abnormal unit. If so, and if selection does not have a cumula- tive effect within a unit, it would be impossible to return to normality from a series of inbred generations of abnormality. But it is possible. Starting with a family which had one normal offspring in a total of 133 (99.2 per cent abnormal) and selecting to reduce the extra veins, the percentage of abnormal offspring in successive generations was 81.8, 66.2, 32.9, 12.5, 17.0, 0.0, 0.0, 0.0, and so on, as a typical normal strain. 14 EXPERIMENTS WITH DROSOPHILA AMPELOPHILA. The building up of an abnormal strain from a long-inbred normal one was also nearly completed when it was stopped by accident. I did not think it worth while to start it anew, as its accomplishment would prove little, since it might be said that the normal strain was a "mixed general population" due to normality (inhibiting) units masking all sorts of latent abnormality units. In nature a small percentage of flies have the fifth longitudinal vein somewhat shortened (see fig. 47) . This variation also appears in the experimental strains. Rather as a matter of curiosity, I selected for shortened veins during a few generations and very quickly obtained such specimens as are illustrated in figs. 48 to 50. One can not go further in this direction without some special technique, because the wings, lacking the support of the veins, droop and catch in the fly's food. Probably breeding could be continued by cutting off the parent wings when matings are made. I did not try it, as it was already very evident that selection was just as effective in the negative as in the positive direction. On the other hand, all attempts to fix, by selection, some particular type of abnormality utterly failed. It was thought possible that the great variety of forms which the extra veins showed was due to a mix- ture of a number of simple forms and that selection might isolate these simple types. The most hopeful was a simple forking of the second longitudinal vein (see fig. 10) . Selection for this type was started sev- eral times, but never went beyond the fifth generation, because, although there were plenty of abnormal flies in each generation, there was no increase in the number showing this particular type, and sooner or later a generation would contain none of them from which to breed. The same was true in the experiments aimed to fix the abnormality on, for example, the third longitudinal vein, but to keep it off of the second. It is easy to have all the abnormal flies abnormal only on the second longitudinal vein, providing one be content with small abnormalities. However, as soon as one increases greatly, by selection, the abnormality on the second vein, the other veins begin to be abnormal. These are the facts: Starting with slight extra veins, either in wild material or in material selected and inbred for normal venation, we can quickly get by selection 100 per cent abnormal offspring. In future generations this strain can be quickly brought back again to its normal condition by selection. Selection also quickly shortens the veins and would probably largely do away with them, provided some technique were adopted to keep the results of selection alive. But selection, accompanied by the strictest inbreeding (brother X sister and parent X child) failed to isolate any unit characterized by a given form or extent of abnormality. The interpretation of these facts would doubtless vary with varying opinions as to unit-characters. INHERITANCE OF ABNORMAL VENATION. 15 THE DATA CONCERNING INHERITANCE. Without reference to the grandparents, the data are summarized in table 7: Table 7. Crosses. Average p. ct, of abnormal offspring. Normal X normal 9.6 35.8 54.7 85.9 Abnormal male X normal female... Normal male X abnormal female... Abnormal X abnormal In this work a fly is counted as abnormal if there is the slightest trace of abnormality in either wing. These results leave no room for doubt concerning the heritability of the tendency toward extra veins. Tables 8 to 19 show the relation between various ancestors and the offspring. The coefficients of association found from these are given below the respective tables. Although these coefficients are greater than expectation on the basis of Pearson's Law of Ancestral Heredity, they do not negative his conclusions. He was very careful to exclude cases in which there is inbreeding or assortative mating. Both were largely practiced in these experiments. These coefficients do show, how- ever, that change of sex in the ancestry does not uniformly weaken in- heritance. Thus, the average coefficient of association between father and sons, and mothers and daughters (no change of sex) is 0.78 ; and that between father and daughters, mothers and sons (one change of Table 8. SONS. Table 9. DAUGHTERS. N. A. <& H B < N. A. N. 2949 837 3786 N. 3011 13-20 4340 A. 1241 1948 3189 A. 951 2703 3714 4190 2785 6975 3962 4092 8054 C. A.=0.694. C. A. =0.736. Table 10. SONS. Table 11. DAUGHTERS. N. A. H 3 o N. A. N. 2801 465 3266 N. 2^83 746 8729 A. 1389 2320 3709 A. 979 3346 43-:.-) 4190 2785 6975 3962 4092 8054 C. A. =0.819. C. A. =0.864 16 EXPERIMENTS WITH DROSOPHILA AMPELOPHILA. 2fc Table 12. SONS. N. A. N. 2813 1079 2892 A. 1329 1705 3034 4142 2784 6926 B Eh Table 13. DAUGHTERS. N. A. N. 2602 1855 4457 A. 1310 2228 3538 3912 4083 7995 C. A. =0.540. C. A.=0.409. 2 ta 6, «=• Table 14. SONS. N. A. N. 2535 4b4 3019 A. 1607 2300 3907 4142 2784 6926 C. A. =0.765. Table 15. DAUGHTERS. N. A. N. 2727 759 3486 A. 1185 3324 4509 3912 4083 7995 C. A. =0.820. -2d S3 Table 16. SONS. N. A. N. 2766 1015 3781 A. 1339 1761 3100 4105 2776 6881 £« Table 17. DAUGHTERS. N. A. N. 2619 1768 4387 A. 1262 2296 3558 3881 4064 7945 C. A. =0.672. C. A. =0.459. Table 18. SONS. N. A. N. 2444 •477 2921 A. 1661 2299 3960 4105 2776 6881 C. A. =0.753. - — aB Table 19. DAUGHTERS. N. A. N. 2630 741 3371 A. 1251 3323 4574 3881 4064 7945 C. A. =0.808. sex) is also 0.78. Considering the grandparents, the average coefficient of association between sons and father's father, and daughters and the mother's mother (no change of sex) is 0.67, while that between sons and the mother's father, and daughters and the father's mother (two changes of sex) is 0.74. This result agrees with that of Blanchard (1903) concerning the coat-color of horses and is not in harmony with Pearson's (1900) and the writer's (1903) concerning the eye-color in man. North Carolina State Library INHERITANCE OF ABNORMAuVftlK^ON. 17 Table 20. — Relation between degree of abnormality in parents and percent- age of abnormal offspring. It will be convenient in the present discussion to adopt the following symbols: Ai denotes a fly that is abnormal in one wing only; A_>, a fly abnormal in both wings, and C. S. (coefficient of symmetry) that per- centage of a given lot of abnormal flies which are abnormal in both wings. Tables 20 and 21 may be summarized as follows: Flies which are so abnormal that both wings are affected not only gave, on the aver- age, a greater percentage of abnor- mal offspring than flies abnormal in only one wing, but the abnormal offspring of the former were more likely to be abnormal in both wings than those of the latter. It must Table 21. — Relation between parents and offspring with respect to one wing or both being abnormal. Parents p. ct of abnormal offspring. Parents. A-2 X -**-2 i Ajx A, 85.1 76.8 71.9 NX A., NX A, N X N P. < t. of abnormal oflbpring. 45.8 351 138 Parents. Offspring. Parents. Offspring. Male. Female. A, A* C. s. Male. Female. A, A2 A2 A2 A2 A, A, A2 A: N A2 A, 268 102 105 149 83 1271 178 118 428 158 0.83 0.64 0.52 0.74 0.66 A, N N N N A2 A, N 55 122 57 285 61 198 53 160 0.53 0.62 0.48 0.36 Table 22.— Relation between parents and offspring with respect to ichich wing is abnormal. [AR=Abnormal in right wing only; -V Abnor- mal in left wing only; \.: Abnormal in both wings; Aj = Abnormal in one wins only.] be noted that this was only "on the average." Although it was an exceptional case, we have seen that all the offspring of mating 405 were normal. The parents each had "great" abnormality in both wings. From table 22 it seems evident that a given asymmetry (abnormal in the right wing only or abnormal in the left wing only) is not inher- ited. The offspring of a parent which is abnormal in the left wing only are as likely to be abnormal in the right as in the left wing, and vice versa. This is in accord with the results obtained by Castle (1906a) for polydactylism of guinea-pigs, Larrabee (1906) for the reversed optic chiasma of fishes, and Priz- bram (1907) for eye-color of cats. As was pointed out, all attempts to fix any particular form of abnor- mality by selection and inbreeding (pure lines?) have failed ; nor has 1 Parents. Offspring. AR A, A, A2 X A2 147 121 +0.10 A.2 X An 58 79 -0.15 A2X AL 56 58 -0.02 36 41 —0-06 30 25 +0.09 A.X A, 23 24 -0.02 NXA, 29 18 +0.23 N x A, 37 28 +0.14 N XN 146 139 +0.02 N XA, 121 106 +0."7 18 EXPERIMENTS WITH DROSOPHILA AMPELOPHILA. there been any apparent tendency to settle down to any definite type. * The "center of disturbance " has remained in the distal portion of the marginal cell closely related to the second longitudinal vein. Next to the second, the third vein has been the most affected, then the fifth; but I have failed to fix the abnormalities in these veins. They are, apparently, all the effects of the disturbing factor or factors, centered on the second longitudinal vein in the marginal cell. The hundreds of families studied showed that it is impossible to pre- dict, from the character of the ancestors, what the form of the abnor- mality will be in the offspring. The most that one can do is to give an approximate estimate of the percentage of abnormal individuals and a still less exact prediction of the average intensity of the abnormalities. THE BEARING OF THESE DATA UPON PROPOSED LAWS OF HEREDITY. In my former paper (1907) I considered that normal venation is more or less dominant over abnormal in the Mendelian sense. Such was the case in the early part of the work, although, as was pointed out, it was the spirit only and not the letter of the law which was followed. When a normal fly, having normal ancestors, was crossed with an abnormal one, practically all the offspring were normal. The abnormalities which did appear were slight, but there was no doubt about their presence. Matings 318 to 322 (see table 36) illustrate such cases. The offspring of matings 347 to 353 are second-generation hybrids from such a cross. They show a condition not very divergent from the Mendelian expecta- tion. Since the number of offspring in most of the families considered here is large, the Galtonian formula can be tested in single families, and it is evidently not at all in accord with the data. Neither is Pearson's modi- fication of it. The fact that normal X abnormal gave, in large families, practically all normal completely negatives for these data all theories which are founded on the hypothesis of equipotency of the two parental characters. On the other hand, while the results of certain matings accord with Mendelian expectation, the fit is far from good in the majority even in the early generations. For instance, we have seen that neither normal nor abnormal breeds true. A Mendelian recessive would be expected to do so ; therefore we can not consider either normality or abnormality to be Mendelian unit-characters in that sense. *Here again (see p. 10) the similarity to the experience of breeders of spotted ani- mals is interesting. Castle (1905), for example, found that "one can by selection progress in either direction through this series of changes, either increasing or de- creasing the number and extent of the pigment patches, but it is impossible without long-continued selection to fix the color-pattern at any particular stage in the series; perhaps it is wholly impossible to do so, as Cuenot (1904) asserts on the basis of his studies on mice, but this I very much doubt." INHERITANCE OF ABNORMAL VENATION. 19 Fig. 52 shows graphically the results of the three sorts of matings : normal X normal, normal X abnormal, and abnormal X abnormal. The first should give one mode at zero abnormality and another at 25 per cent abnormality on the assumption that normality is dominant in the sense in which the term is now used in Mendelian literature. These modes would represent the results of DD X DD and DR X DR, respec- tively. They are present, but the curve runs all the way up to 65 per cent abnormal. The second should give one mode at zero and another at 50 per cent, representing the results of DD X RR and DR X RR, respec- tively. The mode at 5 per cent is marked and might be explained as the 48 46 44 42 40 38 36 34 32 30 ■ Normal x Normal •Normal x Abnormal •Abnormal x Abnormal 0 5 15 25 35 45 55 65 75 65 95 Percentage of abnormal individuals Pig. 52. result of "incomplete dominance, " a thing which is itself badly in need of a Mendelian explanation. At 50 per cent there is a drop in the curve where there should be a mode. There is a strong mode at 75 per cent, where there should be none. This is true both when the male is the normal parent and when the male is the abnormal one (see fig. 53). Abnormal X abnormal should have but a single mode, 100 per cent (or 95 per cent as the figure is drawn), representing the result of RR < RR. Such a mode is pronounced in the curve, being chiefly made up of the families of the abnormal strain after generation vn, but the curve reaches all the way to zero. These data are analyzed in tables 24 to 35, so that there is no need of a further text description of them. They are taken from the early part 20 EXPERIMENTS WITH DROSOPHILA AMPELOPHILA. of the work. The results of the seven crosses between the abnormal and normal strains in the fifty-ninth generation— all that were made at that time— are of interest in this connection and are shown in table 23. Any theory applied to these data must accord with the following facts: (1) Abnormalities occasionally appear in the venation of the wings of wild Drosophila ampelophila. These are usually added veins. Since evolution in the Diptera has been accompanied by a reduction in the number of veins, these abnormalties are of the nature of "reversions. " The tendency to produce extra veins is inherited and has been increased by selection. This is also true of the tendency to shortening of veins. n 9 /\ Male abnormal x Female normal 5> s . \ Male normal x Female abnormal 1' ' \ ° A A \ , \ -' / ^ 1 2 / \ >< >^"* ' V § I •/ V % -K -« * N 2 i — ■ *» O 5 15 25 35 45 55 65 75 85 95 Percentage of abnormal individuals Fig. 63. An examination of more than 50,000 abnormal wings has revealed an immense diversity of forms which the abnormality assumes. Not only are new forms being constantly discovered, but the intensity of the ab- normality has constantly increased as long as selection for that end has been kept up. The limit of the increase was apparently not reached, but the extra veins have always been very crude, only rarely assuming a form and position comparable to ordinary veins. (2) A greater percentage of females than of males is abnormal. The formula Percentage of abnormal sisters = 1.5 X percentage of abnormal brothers approximately describes the average condition in the various families. Attempts to change significantly this relation have failed, and seem destined to fail, for change of sex in the ancestry does not weaken inheritance. (3) The lower range in the variation of abnormality certainly includes barely discernible deviations from normality and presumably just indis- cernible deviations also. The latter would be considered normal. (4) Frequently one wing of a fly is abnormal, the other not visibly so. There is a direct relation between the percentage of abnormal offspring of a given mating which are abnormal in both wings and the total per- centage of abnormal offspring. Furthermore, on the average, parents which are abnormal in both wings give a larger percentage of abnormal INHERITANCE OF ABNORMAL VENATION. 21 offspring than those which are abnormal in one wing only. There is no relation between parents and offspring with respect to the side upon which asymmetrical abnormalities occur. There is a correlation between the two wings of individual flies with respect to the intensity of the abnormality. (5) Normal male X abnormal female gives a greater percentage of abnormal offspring than the reciprocal cross. (6) Not only has the abnormality increased in the abnormal strain, but, in spite of artificial selection to the contrary, an increasingly large number of abnormal individuals appeared for a while in the normal strain and then with the same treatment the percentage again decreased. When the flies are allowed to choose their own mates the percentage of abnormals is kept low even when abnormal flies are added from time to time. (7) Abnormality originally behaved somewhat like a Mendelian reces- sive, but in the later generations departed, in its behavior, very far from that theory as it is now understood. There would be little profit in reviewing the various modifications of the simple Mendelian formula and pointing out in detail why they are not satisfactory in the present case. I have tried most, if not all, of those which have been proposed and also a number of original hypothe- ses involving two or more allelomorphs. All these attempts have been failures with the exception of the idea of variation of potency (Lutz, 1907). If sufficiently elaborated this will "explain" each of the con- ditions set forth above, and until quite recently I believed that the inheritance of the abnormal venation followed this modification of the Mendelian law. It seemed quite probable that there was a single pair of allelomorphs involved — the abnormality-producing factor and its ab- sence — but that the strength of the positive one varied, and that these variations were inherited, making the problem a combination of the inheritance of a fluctuation variate and of Mendelian segregation (Lutz, 1908) . In my report at the time of finishing the work at Cold Spring Harbor I even constructed hypothetical curves for this variation. How- ever, I have since realized that the "explanation " of conditions 6 and 7 was very weak. It was ' ' that in selecting parents to continue the normal strain I merely selected flies having no extra veins. For the most of the time the work of describing offspring was unavoidably so far behind the breeding-work that I did not know what percentage of their brothers and sisters were abnormal. Hence I had no way of judg- ing as to the germinal constitution of the parents. The normal strain is probably a mixture of flies lacking the abnormal factor (might be called NN's) and of flies which have it in hybrid condition of weak allelomorphic strength (NA's). In generations vn to xxn I was prob- ably unconsciously breeding from these NA's. This is an answer to the first part of condition 6. 22 EXPERIMENTS WITH DROSOPHILA AMPELOPHILA. generation. Mating No. Ab- normal. Total. Per cent abnormal. 2561 . ... 2562 43 42 65 47 21 19 14 113 126 130 79 28 27 24 38.1 33.3 50.0 59.5 75.0 70.4 58.3 2629 2630 2631 2633 2645 Total... 251 527 47.6 "We have, now, only to take up the fact that in inheritance these abnormalities follow the spirit but not the letter of the Mendelian law (condition 7) . We might consider that the dominance of normal over abnormal is merely due to the dilution of the abnormality-producing factor in the NA's. If it is strong it may be potent enough to produce abnormalities in spite of this dilu- firm fVmco-ivino-inrwnnlpfpdoTni- Table 23-— Results of seven crosses between tion, tnus giving incomplete aomi abnormal and normal strains in a late nance. Even when it is pure (AA), its fluctuation may give individuals in which the zygotic strength is not great enough to produce abnormalities, thus ac- counting for the normals in the abnormal strain. Whether one could so increase the strength of the abnormality-producing factor that when the selected flies are mated with flies lacking the factor all the offspring will be abnormal is not certain, but table 23 indicates such a possibility. ' ' If, however, we have, in carefully conducted experiments, many flies somatically normal but germinally abnormal, and if by selection it is easy to so weaken the abnormality-producing factor that from a strain 100 per cent abnormal we get and keep one 100 per cent somatically normal (all presumably germinally abnormal, since they came from a 100 per cent abnormal strain) , must we not admit the possibility that all somatically normal flies have the germinal possibilities of abnormality? This makes the problem much simpler, as, leaving out the question of Mendelian segregation, we have only to consider the inheritance of the variations of an abnormality-producing factor, whatever that may be. Let us take up the seven conditions which must be satisfied. Condition 1. — All flies possess the abnormality-producing factor in the germ. It is usually so weak that it has no visible effect upon the soma. Occasionally, however, it is strong enough to do so, and its strength can be so increased by selection that it always does so. Condition 2. —It is necessary to suppose that it takes a greater strength of the germinal factor to have a visible effect upon the male soma than upon the female. This sexual difference of developmental physiology is quite common and the hypothesis will doubtless be readily allowed by most critics in this case. It is interesting to wonder whether the possession of horns by certain male ungulates, while the females lack them, is an extreme example of this same phenomenon. Condition 3. —To be expected on this hypothesis. Condition U— The explanation here would differ according to different notions of the mechanics of heredity. If we accept the apparently most INHERITANCE OF ABNORMAL VENATION. 23 favored notion that there is some specific substance in the germ which produces the character in question and which is divided at the cell- division which separates the substances forming the right side from those forming the left side of the complete soma, it would be difficult to believe the division is always, or in most cases, exactly even. If it is not with respect to the abnormality-producing factor, it would give rise to the phenomenon of asymmetry as to the extent of the abnormality. Only when the factor is rather weak to start with would this deviation from exact equality of division frequently result in the share going to one wing being so small that that wing would be normal while the other wing is abnormal; hence there would be a correlation between the degree of the abnormality and the phenomenon of one wing being normal while the other is abnormal (see p. 5) . The approximate equality of the apportionment of the factor in division may be taken as the expla- nation of the correlation in the intensity of the abnormality in the two wings, and the degree of this correlation is a measure of the degree of equality of the division. Since the going of a slightly greater strength to the right side than to the left, or vice versa, is a mere accident in development, it is not to be expected that there will be an inheritance of a particular side getting the greater strength (see p. 17). But since flies abnormal in only one wing came from germs which had a weak abnormality-producing factor, it is to be expected that the germs they produce will be weak with respect to this factor, and so a smaller per- centage of their offspring will be abnormal than of the offspring of parents abnormal in both wings (see p. 17). Condition 5.— Since it takes a greater strength of the germinal factor to produce abnormalities in the males than in the females, a male somatically normal may be produced by and produce germs containing as strong or stronger abnormality factors than a female which is somat- ically abnormal. Hence, in the long run, normal male X abnormal female will give more abnormal offspring than abnormal male X normal female, because in the latter cross one of the parents (the female) neces- sarily has the factor very weak. Condition 6.— "In selecting parents to continue the normal strain, I merely selected flies having no extra veins. For the most of the time the work of describing offspring was unavoidably so far behind the mating work that I did not know what percentage of their brothers and sisters were abnormal. Hence I had no way of judging as to the ger- minal constitution of the parents." Being unable, by examination of the soma, to tell the exact strength of the germinal content, I uncon- sciously used as parents flies in which the abnormality-producing factor was relatively strong, and thus started and for a time maintained a strain giving a relatively large number of abnormal flies. When the flies were allowed to do their own selecting of mates they were more successful (see p. 36) . 24 EXPERIMENTS WITH DROSOPHILA AMPELOPHILA. Condition 7.— From all we know of the inheritance of fluctuating variations, the mating of one grade with another will give variations about a mid-grade. If it takes a certain grade or strength of the germ- inal factor to produce a somatic effect and we mate a low grade (somati- cally normal) with one just sufficient to produce somatic abnormalities, the variations about the mid-grade will, in most cases, be too low to produce visible abnormalities. In other words, normality will dominate over abnormality. If, however, the parents are such that the grades of the offspring are about that required to produce somatic effects, the dominance will be imperfect. The abnormality will appear to be obey- ing "the spirit of the Mendelian law," but it will naturally "pay little attention to the letter of the simple law or any of the modifying clauses, " especially in the latter generations, where the abnormal strain had the germinal factor much strengthened by selection. It seems to me, then, that if we accept the notion of some specific factor in the germ which brings about the details of the characters of the soma the facts here discussed may be considered to be the result of the action of a factor present in all germs. The strength of this factor varies, and when of a certain strength produces certain visible effects. The partial dominance of normal over abnormal is due to the mean con- dition of the factor in the offspring of (flies with factor strong enough to produce extra veins) x (flies with factor weak) usually being below the strength required to produce abnormality. I have not taken up the resemblance of the behavior of these abnor- malities to that of ' ' ever-sporting varieties, ' ' because I feel that classing these as such would not be a step toward an explanation. It would merely be naming the difficulty. It is also not the intention to imply that this hypothesis would apply to those other cases which are trouble- some from a Mendelian standpoint and to which the principle of vary- ing potency of Mendelian determiners has been applied (for example, Davenport, 1910). It seems not only possible but probable that many apparently non-Mendelian cases may be explained as a combination of alternative and blending inheritance (Lutz, 1908). But a simpler and more probable explanation of these data, provided we accept the some- what dubious "germinal factor " idea, seems to be that we are dealing here solely with a fluctuating character— the strength of the abnormality- producing factor— and that the study of its inheritance is made difficult, if not largely impossible, by the fact that only in the upper part of its range can we judge of the relative values of this variable, for in the lower part its effects are invisible. INHERITANCE OF ABNORMAL VENATION. 25 Table 24.— Percentage of abnormal offspring in 48 matin qa ofNl (ex. Ni X N2) X N9 (ex. Ni X M>). F KM ALES. 0 5 15 25 30 45 55 05 75 95 0 26 8 1 5 4 1 15 1 1 1 25 1 2 1 1 35 45 55 65 75 85 95 Table 25. — Percentage of abnormal offspring in 5 matings of Nt (ex. A6 X A?) X N? {ex. Ai X 4s). FEMALES. 0 5 15 25 35 45 55 65 75 85 95 0 3 5 1 1 15 25 35 45 55 65 75 85 95 26 EXPERIMENTS WITH DROSOPHILA AMPELOPHILA. Table 26.— Percentage of abnormal offspring in 6 matings of NS (ex. At X Ni) X M {ex. A$ X N9). FEMALES. 0 5 15 25 35 45 55 65 75 85 95 0 5 1 15 1 2 25 1 1 35 45 55 65 75 85 95 Table 27.— Percentage of abnormal offspring in U matings of Ns (ex. Ns X Ai) X iV? (ex.Ns X A$). FEMALES 0 5 15 25 35 45 55 65 75 85 95 0 I 5 15 25 35 1 1 45 1 55 65 75 85 95 INHERITANCE OF ABNORMAL VENATION. 27 Table 28. — Percentage of abnormal offspring in 5 matings oj Ns (ex. Ns X N2) X A? (ex. Ns X As). FEMALES. 0 5 15 25 35 45 55 65 75 85 95 0 1 5 1 1 15 25 35 1 45 1 55 65 75 85 95 Table 29.— Percentage of abnormal offspring in 7 mating* of Ns (ex. Ns X N9) X A2 (ex. As X At). FEMALES- 0 5 15 25 35 45 55 65 75 85 1 95 > 0 1 5 15 25 1 1 1 1 35 1 1 45 55 65 75 85 , 95 28 EXPERIMENTS WITH DROSOPHILA AMPELOPHILA. Table 30. — Percentage of abnormal offspring in IS matings of Ns (ex. NsXAl) X A? (ex. NSXA2). FEMALES. 0 5 15 25 35 45 55 65 75 85 95 0 1 1 5 15 25 1 35 1 1 1 45 1 2 1 55 65 2 75 1 85 1 95 1 Table 31.— Percentage of abnormal offspring in 9 matings of Ns (ex. As X As) X 4$ (ex. As X At). FEMALES 0 5 15 25 35 45 55 | 65 75 85 95 0 5 15 25 35 45 1 55 1 65 75 1 1 85 1 95 4 INHERITANCE OF ABNORMAL VENATION. 29 TABLE 32. — Percentage of abnormal offspring in 6 matings of As (ex. As X A9) XN2 (ex. Ns X M>). FEMALES- 0 5 15 25 35 45 55 65 75 85 95 0 5 1 3 1 15 25 35 45 55 65 1 75 85 95 Table 33. — Percentage of abnormal offspring in 12 matings of As (ex. As X N9) X M> (ex. As X Ni). FEMALES. 0 5 15 25 35 45 55 65 75 85 95 0 1 5 1 1 15 1 25 1 35 1 45 55 1 65 1 75 2 1 1 85 95 . 30 EXPERIMENTS WITH DROSOPHILA AMPELOPHILA. Table 34. — Percentage of abnormal offspring in 21 matings of AS (ex. Ns X Ai) X Al {ex. Ns X A9). FEMALES. 0 5 15 25 35 45 55 65 75 85 95 0 1 5 15 25 35 2 1 1 45 2 1 55 1 2 1 65 1 2 75 1 2 85 95 3 Table 35. — Percentage of abnormal offspring in 45 matings of A6 (ex. As X A$) X A2 (ex. As X A$). lXMALES. 0 5 15 25 35 45 55 65 75 85 95 0 5 15 25 35 45 55 1 65 1 75 1 85 1 5 95 36 INHERITANCE OF ABNORMAL VENATION. 31 Table 36. — Percentage of abnormal offspring of 200 families. Oflapring. Parents Mating No. Male. Female. Total Father. Mother. No. recorded. P. ct. abnormal. No. recorded. P.ct ibnormaL ]'. ct abnormal. 211 A 205 A 205 75 30.7 102 64.7 50.3 214 N205 N205 39 30.8 34 47.1 38.4 220 A 207 N207 11 0.0 8 0.0 0.0 226 A 205 A 205 69 46.4 77 80.5 64.4 230 A 205 A 210 38 84.2 27 85.2 84.6 233 A 210 A 210 31 71.0 25 76.0 73.2 238 A 210 A 210 22 36.4 23 52.2 44.4 239 N211 N210 29 51.7 29 69.0 60.3 241 N216 N216 61 0.0 69 0.0 0.0 242 N216 N216 55 0.0 63 3.2 1.7 243 N214 N214 64 28.1 57 36.8 32.2 245 N215 N215 20 20.0 14 42.9 29.4 246 N215 N215 27 25.9 31 32.3 29.3 247 N215 N215 37 29.7 27 40.7 34. 4 253 N220 N220 13 0.0 16 0.0 0.0 257 N211 A 211 87 70.1 90 70.0 70.0 258 N211 A 211 47 93.6 63 96.8 95.5 259 N211 A 211 21 47.6 35 94.3 76.8 273 N242 N242 10 0.0 16 0.0 0.0 276 A 246 N246 18 11.1 23 34.8 24.4 278 N246 N246 16 25.0 17 58.8 42.4 280 N253 N253 10 0.0 20 0.0 0.0 281 N253 N253 37 2.7 52 1.9 22 283 A 257 A 257 47 38.3 54 83.3 62.4 284 A 257 A 257 77 62.3 91 95.6 80.4 285 N257 A 257 72 43.1 72 94.4 68.8 286 A 257 A 257 34 44.1 36 88.9 67.1 288 A 258 A 258 62 54.8 67 95.5 76.0 289 A 258 N258 43 46.5 38 97.4 70.4 291 A 258 A 258 44 65.9 66 93.9 82.7 292 A 258 A 258 16 37.5 27 92.6 72.1 294 N241 N241 16 0.0 19 0.0 0.0 297 N246 N241 34 2.9 57 10.5 7.7 298 N246 A 259 43 48.8 55 76.4 64.3 299 A 238 A 259 26 73.1 28 96.4 85.2 301 N257 N258 53 32.1 51 76.5 53.8 302 N257 A 258 56 51.8 41 92.7 69.1 305 N241 A 233 26 0.0 29 3.4 1.8 316 N276 N225 25 0.0 40 0.0 (Ml 317 N225 A 289 25 16.0 25 4.0 10.0 318 A 288 N225 32 6.3 48 0.0 2.5 319 A 283 N225 30 3.3 42 7.1 5.6 320 A 284 N225 50 10.0 60 20.0 15.6 321 A 284 N225 27 7.4 39 2.6 4.5 322 A 284 N225 48 6.3 53 9.4 7.9 323 N297 N297 29 10.3 35 14.3 12. 5 328 N284 A 284 28 92.9 26 96.2 94.4 330 N284 A 284 48 95.8 66 90.9 930 331 N284 A 284 33 78.8 28 S2.1 80.3 333 A 285 A 285 27 59.3 26 84.6 71.7 338 A 285 A 285 26 53.8 29 62.1 58. 2 339 N285 A 285 34 91.2 45 75.6 82. 3 . i 1 341 N288 A 288 20 90.0 25 92.0 91 1 342 N294 A 284 38 31.6 52 346 33.3 36.6 295 12.4 253 343 344 345 347 N297 N273 N273 N320 A 284 A 284 A 288 N320 35 42 58 32 25.7 14.3 37.9 28.1 47 63 41 55 44.7 39.7 48.8 236 1 32 EXPERIMENTS WITH DROSOPHILA AMPELOPHILA. Table 36. — Percentage of at mormal oj. 7 spring of 200 famil ies — Cont inued. T^O fVknttl Oflspring. Mating No. JrHr* Male. Female. Total p. ct. abnormal. Father. Mother. No. recorded. P. ct. abnormal. No. recorded. P. ct. abnormal. 348 N320 N320 31 19.4 37 21.6 20.6 349 A 320 N320 57 36.8 65 38.5 37.7 350 N320 N320 37 13.5 49 28.6 22.1 351 N320 A 320 36 19.4 32 31.3 25.0 353 N320 N320 37 10.8 24 16.7 13.1 354 A 284 N320 37 45.9 41 63.4 55.1 355 A 284 N321 44 2.3 51 21.6 12.6 356 A 284 N322 38 28.9 30 60.0 42.6 357 N318 N318 32 3.1 35 11.4 7.5 358 A 318 N322 28 0.0 48 2.1 1.3 359 A 318 N322 43 2.3 33 9.1 5.3 364 N317 N317 24 0.0 24 0.0 0.0 365 N345 A 345 44 36.4 52 67.3 53.1 366 N342 A 342 39 30.8 50 42.0 37.1 367 N330 A 330 42 64.3 80 91.3 81.9 368 N330 A 330 43 48.8 55 87.3 70.4 369 N339 A 339 32 34.4 42 76.2 58.1 372 N338 A 338 36 58.3 37 89.2 74.0 375 N344 A 344 29 24.1 36 52.8 40.0 377 N344 A 330 23 60.9 30 93.3 79.2 379 A 349 N349 68 70.6 58 70.7 70.6 380 A 349 N349 26 15.4 29 24.1 20.0 383 A 349 N349 35 25.7 37 59.5 43.1 384 A 349 N349 27 63.0 34 88.2 77.0 387 N368 A 368 32 84.4 36 83.3 83.8 388 A 349 N349 23 4.3 25 12.0 8.3 399 N367 A 367 20 50.0 23 95.7 74.4 401 A 367 A 367 52 92.3 61 93.4 92.9 402 A 367 A 367 103 76.7 90 94.4 85.0 404 A 367 A 367 66 60.6 67 64.2 62.4 405 A 367 A 367 9 0.0 16 0.0 0.0 406 A 367 A 367 56 58.9 76 86.8 75.0 408 A 367 A 367 13 100.0 16 100.0 100.0 409 A 367 A 367 51 43.1 49 93.9 68.0 411 A 367 A 367 23 78.3 39 82.1 80.6 414 N405 N405 56 0.0 69 0.0 0-0 415 N405 N405 75 0.0 74 0.0 0.0 417 N405 N405 38 0.0 73 0.0 0.0 418 A 379 N379 32 71.9 33 75.8 73 8 419 A 379 N379 35 77.1 22 100.0 83.0 433 N399 A 399 24 70.8 17 88.2 78.0 436 A 379 N379 20 75.0 36 88.9 83.9 439 A 408 A 408 54 85.2 53 81.1 83.2 440 A 408 A 408 25 96.0 26 100.0 98.0 453 Nwild A 399 65 1.5 83 2.4 2.0 454 Nwild A 399 49 0.0 52 0.0 0.0 455 Nwild A 399 41 36.6 32 34.4 35.6 456 Nwild A 399 42 4.8 59 10.2 7.9 459 A wild Nwild 49 2.0 59 15.3 9.3 474 N414 N414 18 0.0 15 0.0 0.0 513 A 439 A 439 13 92.3 10 100.0 95.7 514 A 440 A 440 44 97.7 44 97.7 97.7 523 N474 N474 39 0.0 37 0.0 0.0 533 A 513 A 513 44 95.5 54 100.0 98.0 548 A 514 A 514 26 96.2 37 100.0 98.4 557 A 548 A 548 18 94.4 29 100.0 97.9 564 A 533 A 533 4 100.0 4 100.0 100.0 575 N523 N523 39 0.0 20 0.0 0.0 INHERITANCE OF ABNORMAL VENATION. 33 Table 36. — Percentage of abnormal offspring of 200 families — Continued. Iflfepring. Pn rppt^ Mating No. lalrl Male. Female. Total p. ii. abnormal, Father. Mother. No. recorded. P. ct ibnormaL No. recorded. i P.et. ibnormaL 576 N523 N523 52 0.0 48 0.0 ii ii 582 N575 N575 28 0.0 87 0.0 0.0 584 N557 A 557 18 100.0 24 100.0 100.0 585 N576 A 557 48 4.2 45 24.4 14.0 587 N576 N576 29 0.0 21 4.8 20 588 N576 N576 55 0.0 46 0.0 589 A 564 A 564 24 100.0 20 100.0 100.0 590 N588 N582 48 0.0 55 0.0 0.0 591 N588 N588 1 0.0 3 0.0 0.0 592 N587 N582 32 0.0 33 0.0 0.0 593 N587 N582 39 0.0 36 0.0 0.0 594 N585 N585 73 41.1 69 62.3 51.4 600 N588 N588 55 0.0 58 0.0 602 A 589 A 589 21 100.0 25 100.0 1000 605 A 584 A 585 30 40.0 37 67.6 55.2 606 A 584 A 584 2 100.0 7 100.0 100 611 A 585 A"C" 45 17.8 50 38.0 28.4 613 N592 N592 38 0.0 53 1.9 1.1 614 N593 N590 38 0.0 50 2.1 1.1 622 N600 N591 19 0.0 45 4.4 3.1 623 N600 N591 24 0.0 35 0.0 0.0 626 A 602 A 594 53 32.1 60 53.3 43.1 632 A 606 A 606 22 100.0 19 100.0 100.0 633 A 602 A 611 11 90.9 14 92.9 92.0 639 A 594 A 602 1 100.0 7 100.0 100.0 641 A 611 A 605 32 87.5 49 95.9 92.6 642 A 605 A 605 34 55.9 43 651 61.0 643 A 611 A 611 3 100.0 4 100.0 100.0 657 N614 N614 31 0.0 30 0.0 0.0 660 N622 N622 44 2.3 40 5.0 8.6 663 N623 N623 42 0.0 29 0.0 0.0 665 N613 N622 19 0.0 64 0.0 0.0 670 A 626 A 602 35 88.6 45 100.0 95.0 677 A 641 A 626 23 95.7 23 100.0 97.8 678 A 633 A 633 9 100.0 5 100.0 100.0 680 A 642 A 642 34 97.1 49 100.0 98.8 681 A 639 A 639 10 70.0 21 100.0 90.3 682 A 626 A 626 34 91.2 41 100.0 96.0 685 A 643 A 633 18 83.3 23 100.0 92.7 692 A 633 A 633 29 93.1 30 96.7 W.9 694 A 642 A 632 29 75.9 31 83.9 80.0 710 N660 N657 27 0.0 35 0.0 0.0 715 N665 N663 13 0.0 24 0.0 0.0 719 A 670 A 677 7 100.0 11 100.0 100.0 723 A 685 A 682 24 95.8 27 100.0 98. 0 726 A 678 A 680 43 90.7 54 100.0 95. 9 727 A 692 A 692 11 100.0 15 100.0 loi). 0 729 A 681 A 682 . • ■ 1 100. 0 100.0 737 A 694 A 694 9 88.9 7 100.0 93. 8 752 A 726 A 719 45 100.0 56 100.0 1000 753 A 726 A 723 27 92.6 27 100.0 96. 3 763 A 727 A 727 12 100.0 13 11 III II 100. 0 764 A 726 A 727 29 96.6 29 100.0 98 765 A 726 A 729 46 97.8 50 100.0 99 0 767 A 737 A 737 20 80.0 27 10U 0 ;»l .6 785 799 N715 A 753 N715 A 753 63 17 20.6 88.2 83 38 20.5 20.5 90.9 0.0 802 N715 N710 40 0.0 44 0.0 34 EXPERIMENTS WITH DROSOPHILA AMPELOPHILA. Table 36. — Percentage oj abnormal offspring of 200 families— Continued. Offspring Par**1 Mating No. Male. Female. Total Father. Mother. No. recorded. P. ct. abnormal No. . recorded. P. ct. abnormal p. ct. abnormal. 804 N710 N715 19 0.0 19 5.3 2.6 808 A 752 A 765 21 95.2 25 100.0 97.8 813 A 763 A 763 8 100.0 9 100.0 100.0 816 A 764 A 764 26 100.0 23 100.0 100.0 818 A 767 A 767 13 100.0 12 100.0 100.0 831 N804 N802 34 0.0 35 0.0 0.0 833 N802 N802 40 0.0 53 1.9 1.1 836 N785 N785 46 19.6 63 38.1 30.3 851 A 799 A 799 6 100.0 11 100.0 100.0 853 A 816 A 808 25 100.0 41 100.0 100.0 858 A 808 A 808 38 100.0 36 100.0 100.0 880 N836 N831 52 5.8 83 7.2 6.7 882 N833 N836 55 o.o 55 3.6 1.8 886 A 851 A 853 29 100.0 44 100.0 100.0 889 A 858 A 853 16 87.5 35 97.1 94.1 898 N857 N857 55 3.6 63 1.6 2.5 899 N857 N857 50 2.0 33 0.0 1.2 900 A 857 N857 38 2.6 44 13.6 8.5 902 A 886 A 889 28 100.0 36 97.2 98.4 907 A 886 A 886 37 100.0 27 100.0 100.0 908 A 886 A 886 42 100.0 55 100.0 100.0 917 A 886 A 886 39 100- 0 39 100.0 100.0 926 N880 N880 49 6.1 50 6.0 6-1 XXII A 886 N882 20 65.0 27 74.1 70- 2 943 A 902 A 902 59 98.3 74 100.0 99.2 946 A xxn A xxn 44 40.9 62 62.9 53.8 947 A xxn N xxn 50 52.0 78 66.7 60.9 948 N xxn N xxn 54 25.9 59 40.7 36. 9 953 N926 N926 40 0.0 58 13.8 8-2 954 N926 A 908 43 32.6 46 58.7 46.1 955 N926 A 908 40 27.5 53 26.4 26.9 967 A XXII A xxn 37 54.1 46 71.7 63.9 968 A xxn A xxn 23 87.0 32 100.0 94.5 970 N xxn N xxn 44 31.8 43 34.9 33.3 983 N946 N946 46 39.1 43 60.5 49.4 984 A 946 A 946 33 81.8 51 90.2 86.9 985 A 946 N946 41 56.1 41 78.0 67.1 986 N947 N947 29 24.1 36 389 32.3 989 A 947 N947 27 66.7 29 82.8 75.0 997 A 943 | A 943 27 77.8 28 85.7 81.8 1006 A 930 N957 18 55.5 20 60.0 57.9 1017 N953 N953 7 0.0 8 12.5 7 3 1021 A 947 N947 21 47.6 22 45.5 1 ■ o 46.5 1064 1069 N997 N1017 A 997 N1017 30 24 50.0 0.0 38 40 78.9 2.5 66.2 1 6 1071 A 981 N1006 4 0.0 7 14 3 9.1 0.0 35.1 32.2 0.0 0.0 9.3 1.4 4.1 5.4 12.5 17.0 15.2 2.2 10.7 1111 N1062 N1069 9 0.0 9 0 0 1117 N1064 N1064 19 26.3 18 44 1118 1123 N1064 N1069 N1064 N1069 55 52 23.4 0.0 63 65 39.7 0.0 0 0 1124 N1069 N1069 19 0.0 36 1141 1153 2 A 1102 Nllll 3 N 1071 Nllll 39 69 10.3 0.0 58 74 8.6 2.7 6.4 5.7 16.7 20.7 21.4 1.4 13.6 1154 1156 1158 1166 1172 1176 1177 Nllll Nllll N1117 N1158 N1156 N1154 N1154 Nllll Nllll N1118 N1158 N1156 N1156 N1156 51 39 2 42 37 70 53 2.0 5 1 0.0 11.9 8.1 2.9 7.5 47 53 6 58 42 69 59 INHERITANCE OF ABNORMAL VENATION. 36 TABLE 36. — Percentage of abnormal offspring of 200 families — Continued. P'l »»£in4o Offspring. Mating No. l rtl tr Male. female. L.tal Father. Mother. recorded. P.ct abnormal. recorded. 1-. d abnormal. p. ct ibnorm&L 1181 A 1141 4 N 1141 39 10.3 70 J2.9 18.3 1190 2 A 1141 4 N 1141 45 11.1 53 11.3 11.2 1192 N1177 N1177 39 2.6 28 17.9 !» 0 1193 N1177 N1177 49 2.0 73 13 1.6 1197 N1176 N1176 61 0.0 53 1.9 0.9 1204 N1166 N1166 2 0.0 6 o.o 0.0 1208 A 1181 3 N 1181 49 8.2 53 11.3 9.8 1213 N1166 N1166 33 3.0 17 23.4 15. 0 1217 3 A 1190 4 N 1190 28 21.4 29 114 31.6 1221 N1197 N1197 43 4.7 47 64 5.6 1226 N1193 N1193 29 0.0 21 9.5 4.0 1229 N1204 N1204 5 0.0 11 00 0.0 1254 N1229 N1229 23 0.0 21 0.0 0.0 1256 N1226 N1226 39 2.6 46 10.9 7.1 1266 N1254 N 1254 25 4.0 26 3.8 39 1268 N 1254 N1254 13 0.0 12 0.0 0.0 1269 N1254 N1254 4 0.0 9 0.0 0.0 1282 N1256 N1256 4 0.0 4 0.0 0.0 1286 N1269 N1269 27 0.0 29 0.0 0.0 1293 N1286 N1286 55 1.8 53 5.7 3.7 1300 N1282 N 1282 20 0.0 24 0.0 0.0 1309 N1293 N 1293 12 0.0 14 0.0 1327 N1300 N1300 52 0.0 48 0.0 00 1347 N1327 N1327 46 0.0 59 0.0 0.0 1394 N1347 N1347 61 0.0 73 0.0 0.0 1428 3 N 1394 3 N 1394 62 1.6 66 0.0 0.8 1454 N1428 N1428 65 0.0 64 0.0 0.0 1476 N1454 N1454 43 0.0 46 0.0 0.0 1533 N1476 N1476 34 0.0 33 0.0 0.0 1565 N1533 N1533 64 0.0 74 0.0 0.0 1587 * A 1498 * A 1498 37 24.3 40 15.0 19.5 1588 * A 1492 * A 1492 44 2.3 44 0.0 1.1 1626 N1565 N1565 25 0.0 32 0.0 0.0 1668 A 1587 A 1588 43 2.3 69 2.9 2.7 1687 A 1587 N1587 55 1.8 63 12..7 7.6 1701 N1626 N1626 77 0.0 68 0.0 0.0 1722 N1668 N1668 30 0.0 70 4.3 3.0 1785 N1687 A 1687 29 10.3 33 152 12.9 1787 N1701 N1701 32 0.0 40 0.0 .0 1A /* 1788 A 1668 N1668 5 0.0 80 11.3 10.6 1830 N1787 N1787 45 0.0 sS 0.0 0.0 *7 O 1857 N1722 N1722 20 10.0 62 6.4 1 t \ 1859 A 1785 N1785 21 0.0 30 33.3 19. b 0.0 12.9 0.0 9.7 0.0 31 T 21 28 5 8 0.0 0.0 '.0 1890 N1830 N 1830 7 0.0 8 0.0 1961 N1859 A 1859 49 16.3 44 9.1 2006 N1788 A 1788 18 5.6 40 25.0 2013 N1890 N1890 12 0.0 16 0.0 2084 A 1857 A 1857 8 12.5 23 8.7 0.0 39.3 33.3 30.0 0.0 43.8 0.0 0.0 90 0.0 100.0 2194 N2013 N2013 18 0.0 9 2224 2226 2296 2313 2366 A 1961 A 1961 A 2084 N2194 A 2296 A 2006 A 2006 A 2084 N2194 A 2224 13 68 25 14 8 15.4 10.3 28.0 0.0 50.0 28 99 10 23 16 2367 N2313 N2313 11 0.0 15 01 2458 2471 N2367 A 2366 N2367 A 2366 24 78 0.0 79.5 21 114 46 '.1 2503 2524 N2458 A 2471 N2458 A 2471 53 45 0.0 100.0 •From wild material (see page 13). THE EFFECT OF SEXUAL SELECTION.* It is relatively easy to get by artificial selection a strain of Drosophila ampelophila in which practically all the individuals possess extra wing- veins. Also, by selection one can reduce the amount of venation. The latter strain is manifestly not fitted to maintain itself, because the wings, deprived of the support of the veins, droop and catch in the food of the insect, resulting in the insect's death. On the other hand, the wings of the extra- veined race are strong, the individuals are vigorous and fer- tile. What would be the fate of such a race if turned loose in nature (a) where they would find plain-winged individuals with which to breed and (6) where they were isolated from plain-winged individuals? Rea- soning from the fate of most feral domestic races, one would expect that in the former case they would soon disappear, although the reason assigned for their disappearance would be the vague one that they would be ''swamped." In the latter case many would expect them to keep the domestic characteristics. Two cubic feet of space and a few decaying bananas form conditions sufficiently feral for the purpose of testing what would happen. On May 2 I released 'in a large battery-jar an equal number of flies from one of my extra-veined strains and from one of my plain-winged strains. This would clearly give the extra-veined an advantage, for not often will a new form make up 50 per cent of the population. On May 19 only 26 per cent of the flies in the jar showed extra veins and these veins were not as pronounced as those of the original 50 per cent. By May 26 the number was reduced to 11 per cent. It was 7 per cent on June 9, and two weeks later (June 23) only 1 per cent showed any trace of extra veins. On February 19 I released in a similar jar a population of flies selected from an extra-veined race on the basis of well-developed extra veins. No plain-winged flies were introduced. However, after six weeks (March 31) only 93 per cent showed extra veins and in none of these cases were the extra veins very strong. On April 24 there were only 84 per cent; May 28, 72 per cent; June 23, 49 per cent; and by August 3 only 5 per cent showed any trace of extra veins. As has been shown, plain-winged individuals occasionally turn up in carefully-bred extra-veined races, but it was, at first, puzzling to see how these occasional "reversions" could get such a foothold as to supplant the extra- veined flies which were in the j ar by the hundreds. The expla- nation was found while testing the selective value of the prominent male secondary sexual character on the anterior tibiae— the large tibial comb. *Paper read before the American Society of Naturalists, Boston meeting, 1909. 36 THE EFFECT OF SEXUAL SELECTION. 37 I cut them off of a plain-winged male and left them on a male of the extra-veined race and vice versa. These two males were then given to a female as mates. By a study of her offspring I could tell, in a rough way, which mate she preferred. To my surprise she chose almost exclu- sively the normal male, whether he had tibial adornments or not. Then, without removing the tibial combs, I gave plain-winged and extra-veined individuals the choice between mates which, as far as I could determine, were alike in all particulars such as age, nutrition, activity, and time since last copulation, but differed in that one had extra veins while the other had not. I watched each experiment until copulation had taken place. When the extra venation in one mate was great, the chooser, whether male or female, normal or extra-veined, chose the normal mate. I then tried weaker degrees of the character and in 85 experiments, mostly with flies having the extra veins only very slightly developed, 61 of the choices were in favor of the wild type. The basis upon which these flies discriminate against extra-veined individuals when choosing a mate is a matter for further study. There is an elaborate "courtship," in which the flirting of the wings in front of the prospective mate plays a large part. It seems as though a choice were made on the basis of sight, but I doubt whether that is the case. However, there is no doubt of the choice. It is a clear case of the undoing of artificial selection by sexual selection. DISUSE AND DEGENERATION. * One of the several much-discussed but little-tested problems of the theory of evolution is that of the inherited effects of disuse. I believe that there is a pretty general idea that when a species no longer has need for an organ that organ will degenerate. The explanations of this degener- ation are varied, but the most popular seem to be the inheritance of acquired characters, panmixia and selection. It is indisputable that in the life of an individual many unused organs do degenerate, but it is far from proven or even satisfactorily indicated that this ontogenetic degen- eration is followed by a phylogenetic degeneration. There is no doubt that many degenerate organs are not used in any way; but who can say whether this disuse has preceded degeneration as a cause or merely fol- lowed as a necessary consequence ? Before attempting to explain the phylogenetic degeneration which follows disuse it seems desirable to find a clear case of such a sequence, and this quest was the purpose of the experiment with Drosophila ampelophila upon which I wish briefly to report. These insects are normally very good fliers, possessing wings which are relatively quite large. In my experiments, however, they were con- fined in glass vials barely large enough to contain the food. The only opportunity they had to fly was when they were transferred from one vial to another. This was done only three times a week. Such flight could at most not be more than 5 cm., and was, as a matter of fact, rarely made, as they usually walked. The experiments are complicated by several facts which must be con- sidered. These fall into two groups: First, those which might explain the absence of degeneration in the wings. Disuse does not affect, during the life of an individual, the wing-dimensions, for after an insect's wings are expanded there is no change in them and, of course, they are not subject to the effects of use and disuse before they are expanded. However, the degeneration of beetle- wings when the elytra are fused, of the wings of cave insects, of parasites, and of the wings of many female Lepidoptera are used as stock examples of disuse. Furthermore, if there be anything in the theory of hormones (of which Cunningham has recently made so much) or the various forms of the memory theory of inheritance, we would expect phylogenetic degeneration because of the germ-plasm receiving the news that the wings are not being used, providing the plasm is in condition to *Paper read before the American Society of Zoologists, Baltimore meeting, 1908. 38 DISUSE AND DEGENERATION. 39 receive and act upon such a stimulus. In certain insects the germ-cells are all practically matured before or by the time the wings are expanded and ready for use. This, however, is not the case with Droaophila. Not only are the germ-cells not all matured by the time it becomes adult, but they are in all stages of development and continue to mature, a : at a time, for a month thereafter. In these experiments I rarely u as parents the individuals coming from first-laid eggs, so that there were strong chances of my using affected germ-plasm if such exists. Any experiment, such as this, is always open to the criticism that it has not been sufficiently long continued, but I am sure that most will agree that 43 generations, combined with microscopic measurements and the delicacy of biometric analysis, ought to give a satisfactory indication of what is taking place. The second set of considerations might explain any observed degen- erations without reference to the disuse. Excessive inbreeding was practiced, sister usually being bred to brother. This was necessary f < >r, if I had planned to stop at this point and had wished to entirely avoid inbreeding, I would have needed more than 8 trillion flies with which to start the work. Inbreeding is supposed to lead to degeneration and might thus be solely accountable for degeneration, or it might assist disuse. Unnatural conditions might have adversely affected the flies. Confine- ment itself, apart from the entailed disuse, might at least help to bring about degeneration. Furthermore, I kept the insects breeding winter and summer, with no rest for hibernation and with no change of food. There was no conscious selection favoring perfect and large wings, as all measurements of this strain were made quite recently and the vari- ations in wing-dimensions are not readily appreciable, hence the removal of selection in favor of good wings might result in panmixia and conse- quent degeneration. Finally, I was constantly on the lookout for signs of degeneration, as I hoped and still do hope to produce a wingless Dro- sophila. My desire might have influenced my actions and an unconscious selection on my part might have reduced the size of the wings without disuse playing a part. The only necessary answer to this second set of considerations is that, in spite of the possibility of the degenerating effect of disuse being helped by inbreeding, unnatural conditions, panmixia, or selection, there has been no degeneration. Evidence of degeneration was sought for by carefully measuring the expanded wings of the individuals belonging to successive stages of the experiment. In making these measurements one may not mix the sa because of the sexual difference in size. Therefore the females al< were used, since among insects it is more commonly the females which have degenerate wings. The results are shown in table 37. where units of length equal 1 mm. 40 EXPERIMENTS WITH DROSOPHILA AMPELOPHILA. Table 37. — Mean wing-dimensions at various periods of continued disuse. Generations of disuse. Length of wing. Breadth of wing. Length X breadth. 1st to 3rd 67.00 ±0.12 66.50 ±0.11 64.91 ±0.12 67.67 ±0.18 32.55 ±0.07 33.60 ± 0.06 31.98 ±0.06 34.06 ±0.08 2223.12 ± 8.85 2281.12 ± 7.41 2115.44 ± 7.42 2358.94 ±11.46 17th to 19th 33rd to 35th 41st to 43rd If the experiment had stopped at the end of the thirty-fifth generation it would have appeared from this table that the wings were actually- getting smaller, since the area, as judged by length X breadth, was smaller in the second lot than in the first, and still smaller in the third — the difference being nearly ten times the expected error. However, this would have been a hasty conclusion. The fourth lot is as much larger than the first as the third is smaller. So we must conclude that there is no evidence that the constant disuse of the wings during more than 40 generations has had any effect. BIBLIOGRAPHY. Blanchard, N. 1903. On the inheritance in thoroughbred horses. Biometrika, II, 229-234. Castle, W. E. 1905. Heredity of coat characters in guinea-pigs and rabbits. Carnegie Institu- tion of Washington, Publication No. 23. 1906a. The origin of a polydactylous race of guinea-pigs. Carnegie Institution of Washington, Publication No. 49. 19066. Inbreeding, cross-breeding, and sterility in Drosophila. Science, n. s. , xxm, No. 578. (See also Proc. Am. Acad., xli, No. 33.) Cuenot, L. 1904. L'he're'dite de la pigmentation chez les souris. ( 3me note. ) Arch. Zool. Exper. et Gen., ser. 4, torn. 2, Notes et Revue, pp. xlv-lvi. Davenport, C. B. 1910. The imperfection of dominance and some of its consequences. Amer. Nat. xliv, No. 519. Jennings, H. S. 1908. Heredity, variation, and evolution in Protozoa. II. Proc. Am. Phil. Soc, xlvii, No. 190, p. 523. Larrabee, A. P. 1906. The optic chiasma of Teleosts : A study of inheritance. Proc. Am. Acad. xlii, No. 12. Lutz, F. E. 1903. Note on the influence of change in sex on the intensity of heredity. Bio- metrika, II, pp. 237-240. 1907. Paper read at the Seventh International Zoological Congress. 1908. Combinations of alternative and blending inheritance. Science n s xxvm, No. 714. Pearson, Karl. 1900. On the inheritance of characters not capable of exact quantitative measure- ment. Phil. Trans., A., cxcv, pp. 115-117. Prizbram, Hans. 1907. Vererbungsversuche uber asymmetrische Augenfarbung bein Angorakat- zen. Archiv f. Entwicklungsmechanik, xxv, 260-265. North Carolina State Library Raleigh PAMPHLET BINDER" ■ Syrocuse, N. Y. ^^Z Stockton, Calif. NorthCaro.inaStateUniversityUbrar.es S02776108 G PP *ta