the results are, I think, of considerable interest from the aspect of heredity. In the following summaries are given the results of various breedings carried out in 1929 and succeeding years.

Pair “A” (original parents).

Alucita monospilalis ♂ x Alucita monospilalis♀.

Progeny (first filial generation).

monospilalis, 9 ♂ ♂ + 7 ♀ ♀. lycosema, 3 ♂ ♂ + 1 ♀. intermediate, 3 ♂ ♂ + 4 ♀ ♀. Attained maturity, 15 ♂ ♂ + 12 ♀ ♀ = 27.

From the above progeny, constituting the first filial generation, the pairs “A1,” “A2,” “A3,” “A4” (all monospilalis ♂ x monospilalis ♀), “A5” (lycosema ♂ x lycosema ♀), “A6” (lycosema ♂ x monospilalis ♀), and “A7” (lycosema ♂ x intermediate ♀) were taken.

Pair “A1” (from first generation; second series parents).

A. monospilalis♂ x A. monospilalis ♀.

Progeny (second filial generation of “A”).

monospilalis, 6 + 4 (sp.) = 10♂♂; 23 + 7 (sp.) = 30 ♀ ♀.

lycosema, 0.

intermediate, 3 + 6 (w) = 9♂♂.

Attained maturity, 19♂♂ + 30 ♀ ♀ = 49.

Eggs laid, 100; eggs hatched, 100.

Mortality, 51 = 51% (all as young larvae).

This pair was exceptionally prolific.

From this second generation another monospilalis♂ x monospilalis ♀ pair, “A1A1,” was taken and also an intermediate♂ x monospilalis ♀ pair, “A1A2.”

Pair “A1A1” (second generation; third series parents).

A. monospilalis♂ x A. monospilalis ♀.

Progeny (third filial generation of “A”).

monospilalis, 3♂♂; 17 ♀ ♀.

lycosema, 14♂♂; 8 ♀ ♀.

intermediate, 2 + 2 (w) = 4♂♂; 1 + 1 (s) + 2 (w) = 4 ♀ ♀.

Attained maturity, 21♂♂ + 29 ♀ ♀ = 50.

Eggs laid, 54; eggs hatched, 51.

Mortality, 4 = 7.40% (3, equalling 5.55%, failed to hatch; 1, equalling 1.85%, died as mature pupa).

From this third generation yet another monospilalis♂ x monospilalis ♀ pair, “A1A1A1,” was taken and the breeding continued.

Pair “A1A1A1” (third generation; fourth series parents).

A. monospilalis♂ x A. monospilalis♀.

Progeny (fourth filial generation of “A”).

monospilalis, 0♂♂; 6♀♀.

lycosema, 0.

intermediate, 2 + 3 (w) = 5♂♂; 1 (w) = 1♀.

Attained maturity, 5♂♂ + 7♀♀ = 12.

Eggs laid, 12; eggs hatched, 12.

Mortality, 0.

Rising also from the progeny of pair “A1” we have the intermediate♂ x monospilalis ♀ pair, “A1A2,” contemporaneous with pair “A1A1.”

Pair “A1A2” (second generation; third series parents).

intermediate♂ x monospilalis ♀.

Progeny (third filial generation of “A”).

monospilalis, 3♂♂; 11 ♀ ♀.

lycosema, 0.

intermediate, 7 + 4 (w) = 11♂♂; 1 (w) = 1 ♀.

Attained maturity, 14♂♂ + 12 ♀ ♀ = 26.

Eggs laid, 34; eggs hatched, 32.

Mortality, 8= 23.52% (2, equalling 5.88%, failed to hatch; 1, equalling 2.94%, died as half-grown larva; 5, equalling 14.70%, died as pupae).

Reverting to the pairs derived from the same source as “A1” we have the following, “A2,” “A3,” “A4,” “A5,” “A6,” and “A7.” It will be remembered that “A1” was monospilalis♂ x monospilalis ♀.

Pair “A2” (first generation; second series parents).

monospilalis♂ x monospilalis ♀.

Progeny, 0. This pair unprolific.

Attained maturity, 0.

Eggs laid, 1; eggs hatched, 1.

Mortality, 1 = 100% (died as young larva).

Pair “A3” (first generation; second series parents).

monospilalis♂ x monospilalis ♀.

Progeny (second filial generation of “A”).

monospilalis, 2♂♂; 13 + 1 (sp) = 14 ♀ ♀.

lycosema, 0.

intermediate, 12 + 6 (w) = 18♂♂.

Attained maturity, 20♂♂ + 14♀♀ = 34.

Eggs laid, 36; eggs hatched, 36.

Mortality, 2 = 5.55% (died as young larvae).

Pair “A4” (first generation; second series parents).

monospilalis♂ x monospilalis♀.

Progeny (second filial generation of “A”).

monospilalis, 1♀.

lycosema, 0.

intermediate, 1♂.

Attained maturity, 1♂ + 1♀ = 2.

Eggs laid, 4; eggs hatched, 4.

Mortality, 2 = 50% (died as young larvae).

Pair “A5” (first generation; second series parents).

lycosema♂ x lycosema♀.

Progeny (second filial generation of “A”).

monospilalis, 1 (sp) = 1♂; 1♀.

lycosema, 0.

intermediate, 2♂♂.

Attained maturity, 3♂♂ + 1♀ = 4.

Eggs laid, 4; eggs hatched, 4.

Mortality, 0.

Pair “A6” (first generation; second series parents).

lycosema♂ x monospilalis♀.

Progeny (second filial generation of “A”).

monospilalis, 1♂; 2♀♀.

lycosema, 0.

intermediate, 7 (s) = 7♂♂; 3 (s) + 4 = 7♀♀.

Attained maturity, 8♂♂ + 9♀♀ = 17.

Eggs laid, 18; eggs hatched, 18.

Mortality, 1 = 5.55% (died as mature larva).

Pair “A7” (first generation; second series parents).

lycosema♂ x intermediate ♀.

Progeny (second filial generation of “A”).

monospilalis, 0♂; 5 ♀ ♀.

lycosema, 8♂♂; 5 ♀ ♀.

intermediate, 2 + 3 (s) = 5♂♂; 2 + 1 (s) = 3 ♀ ♀.

Attained maturity, 13♂♂ + 13 ♀ ♀ = 26.

Eggs laid, 29; eggs hatched, 29.

Mortality, 3 = 10.34% (2, equalling 6.89%, as half-grown larvae; 1, equalling 3.44%, as mature pupa).

In addition to the preceding, further breeding was done with pairs “B” (monospilalis♂ x lycosema♂) and “C” (lycosema♂ x lycosema♂).

Pair “B” (original parents).

A. monospilalis♂ x var. lycosema♀.

Progeny (first filial generation).

monospilalis, 1♂; 3 ♀ ♀.

lycosema, 3♂♂; 6 ♀ ♀.

intermediate, 4 + 7 (w) = 11♂♂; 1 + 1 (w) = 2 ♀ ♀.

Attained maturity, 15♂♂ + 11♂♂ = 26.

Eggs laid, 26; eggs hatched, 26.

Mortality, 0.

From the above progeny, constituting the first filial generation of pair “B,” was taken the pair “B1,” again monospilalis♂ x lycosema♀.

Pair “B1” (first generation; second series parents).

monospilalis♂ x lycosema♀.

Progeny (second filial generation of “B”).

monospilalis, 1♂; 19 + 1 (sp) = 20♀♀.

lycosema, 0.

intermediate, 4 + 5 (w) = 9♂♂; 1 (w) = 1♀.

Attained maturity, 10♂♂ + 21♀♀ = 31.

Eggs laid, 38; eggs hatched, 38.

Mortality, 7= 18.42% (1, equalling 2.63%, died as young larva; 6, equalling 15.78%, died as pupae).

Pair “C” (original parents).

var. lycosema♂ x var. lycosema ♀.

Progeny (first filial generation).

monospilalis, 0♂♂; 1 ♀ ♀.

lycosema, 0.

intermediate, 0.

Attained maturity, 0♂♂ + 1 ♀ ♀ = 1.

Eggs laid, 2; eggs hatched, 2.

Mortality, 1 = 50% (died as young larva).

In the preceding data of the various pairs the abbreviations mean as follows:—(w) weak, (s) strong, (sp) strongly spotted. A weak intermediate is a form more closely approaching the pure monospilalis, while a strong intermediate approaches more closely to the other extreme of the varietal range, the var. lycosema. A strongly spotted form of monospilalis is, of course, approaching the intermediate condition.

find pure lycosema characters present in the first filial generation (14.81%), entirely absent in the second, recurring in the third (44%), and again absent in the fourth filial generation. However, in both the second and fourth generations the intermediates, present in the strengths of 40.43% (average) and 50% respectively, indicate a trend towards the lycosema variety.

In the progeny, the third filial generation of “A,” of pair “A1A2” (intermediate♂ x monospilalis ♀) there is an absence of pure lycosema. Having regard to the intermediate male parent, to the trend exhibited by the progeny of pure monospilalis parentage, and to the presence of lycosema in the progeny, also the third filial generation of “A,” of pair “A1A1,” I had expected some pure lycosema characters in the offspring of “A1A2.” Instead, 53.84% displayed the coloration characteristics of the female parent and 46.15% displayed those of the male parent.

Pair “A5” (lycosema♂ x lycosema ♀) gave rather a reversal to the results of the other breedings inasmuch as the characteristics of the parents did not appear in pure form in any of the progeny. In this case 50% of the offspring were pure monospilalis and 50% were intermediate.

Pair “A6” (lycosema♂ x monospilalis ♀) produced no pure lycosema, but intermediates (82.35%) were numerically superior among the progeny.

On the other hand, pair “A7” (lycosema♂ x intermediate♂) produced 19♀23% pure monospilalis, 50% pure lycosema, and 30.76% intermediates.

The inheritance-diagram of whole “A” series is:—

Working with pairs having no relationship with pair “A,” we find that pair “B” (monospilalis ♂ x lycosema♀) yielded, in the first filial generation, 15.38% pure monospilalis, 34.61% pure lycosema, and 50% intermediates. In the second filial generation pair “B,” per medium of pair “B1,” gave 67.72% monospilalis, 32.25% intermediates, but no lycosema.

The inheritance-diagram of whole “B” series is:—

Pair “C” (lycosema♂ x lycosema ♀) gave results similar to those given by pair “A5,” inasmuch as none of the progeny displayed the pure parental characteristics, 100% being pure monospilalis.

An interesting feature regarding the breeding of the varieties of this species is that under both natural and artificial conditions like seems to prefer to mate with like, i.e., monospilalis to monospilalis, lycosema to lycosema, and intermediate to intermediate. Cross-mating (lycosema to monospilalis, etc.) does occur, but not commonly, even though the varieties should be present in roughly equal numerical strength. Under artificial conditions, in order to be sure of cross-mating, I found it necessary to segregate the individuals I desired to mate. If I merely left a number of individuals in a cage and segregated them as they paired, I usually found myself with like-to-like matings only.

Besides the instances of cross-mating, under natural conditions, which I have observed myself, the only other record I have been able to find is in Hudson (1928); he records a single instance of such a mating.

Regarding the cross-matings, in pairs “A1A2,” “A6,” “A7,” “B,” and “B1,” it will be seen that the results, in so far as they go, differ, in spite of the distinctiveness of the parents constituting each pair, from the usual principles of Mendelian inheritance.

The usual principle of Mendelian inheritance may be expressed diagrammatically thus:—

“T” being the tall parent and “S” the short. The tallness is the dominant characteristic and the shortness the recessive. “T(S)” indicates that the recessive characteristic is present, latent in some of the germ-cells, but unexpressed in development.

Taking the cross-mated pairs in order, expressing the results diagrammatically, and comparing them with the preceding T. x S. diagram, we get the following. The bracketed numbers indicate the numerical strengths of the varieties:— Pair “A1A2”—♂

In this case the characteristics of neither parent proved dominant. Thus, instead of a latent recessive character, we find the features of each parent fully developed and almost evenly divided in the progeny.

Pair “A6”—

Here the pure lycosema characteristic becomes recessive and unexpressed, but intermediate features make their appearance in numbers far in excess of those of the other parent. Thus the bulk of the progeny in the first generation may be said to combine and express the parental characteristics, the extreme features being toned down in the course of the combination. Pair “A7”—

In this case, as in “A1A2,” neither parental characteristic becomes recessive and unexpressed in the first generation; lycosema constitutes 50% of the progeny and the other parent is expressed in development to a moderate degree. But a third factor, monospilalis, appears—it does not resemble either parent nor does it combine the parental qualities as does “I” in “A6.” Pair “B”—

Again the characteristics of both parents are expressed in the first generation and again an outside factor enters. This third factor may be said to combine and tone down, or achieve a compromise