Linoleic Acid, Vitamin E and Other Fat-Soluble Substances in the Nutrition Of Certain Insects, Ephestia Kuehniella, E. Elutella, E. Cautella and Plodia Interpunctella (Lep.)

The nutritional requirements of the moths in question are fully described in a previous publication (Fraenkel & Blewett, 1946b) which also contains full details about the methods of computing the diets, breeding and testing. The basal diet used in all tests with E. kuehmella consisted of

The diet used for E. elutella, E. cautella and Plodia interpunctella was the same, except that it contained 50 parts of casein and 50 parts of glucose.

The quantity of wheat-germ oil in the diets was usually of the order of 1 %, or slightly less, of the dry diet, except where otherwise stated. The saponifiable fraction of wheat-germ oil was supplied in the same quantities, and the unsaponifiable fraction was about of the dry diet. For the saponification we followed the instructions of Hilditch( 1940), with only slight modifications. 20 g. of oil were saponified for 3 hr. in a solution of 5 g. KOH in 100 g. 95 % alcohol, the soaps extracted for 24 hr. with ether in continuous extractor, acidified, and the fatty acids shaken out with ether. The ‘unsaponifiables’ were submitted to a second saponification for 6 hr. and collected by continuous extraction with ether.

All quantities are expressed per g. of the dry diet, except where otherwise stated. As in previous publications, many of the results are graphically summarized in curves in which the total number of pupae formed is plotted against time. Each figure represents results obtained in one set of tests.

If the artificial diet described above also contains wheat-germ oil, Ephestia kuehmella grows almost as fast on it as on wholemeal flour. In the absence of wheat-germ oil growth is very slow and the mortality often high. In the tests recorded in Fig. 5, for instance, only three out of twenty larvae pupated after a long delay and all the moths failed to emerge from the pupal cases. With E. elutella the result is slightly different. In the absence of wheat-germ oil growth is also impaired but sometimes to a lesser degree, and most larvae eventually pupate. Again, moths never emerge from these pupae (Fig. 6 and Fraenkel & Blewett, 1946b, Fig. 5).

An examination of old pupae which failed to emerge invariably showed the presence of fully developed moths inside the pupal cases; they are pigmented and in the state in which they would normally emerge. They are alive if examined on the 12th or 13th day after pupation, the normal duration of the pupal stage at 25° C., but die within 1 or 2 days. In some cases there were signs of an attempted emergence, such as a burst head capsule and the freeing of parts of the head, the proboscis or the anterior legs, and occasionally a moth emerged in a malformed state. With insufficient wheat-germ oil moths emerge with wings partly or entirely lacking in scales. In a series of diets with graded doses of wheatgerm oil all stages between failure to emerge and normal moths are produced. With E. kuehmella the moths are normal with 4·3 mg. wheat-germ oil/g. of the diet. Lowering the amount of wheat-germ oil has the effect of producing moths with increasingly naked wings. For the purpose of classifying degrees of deficiency we are distinguishing the following eight classes, illustrated in Fig. 1 :

The effects of varying amounts of wheat-germ oil in the diet of E. kuehmella are shown in Table 1 (1 a–e).

The question here arises as to the fate of the missing scales. These are found stuck on to the inside of the empty pupal cases where they are seen as dark regions after the moth has emerged. It appears that the formation of scales proceeds normally up to the time of emergence of the moth. Then during emergence, instead of separating from the pupal case and remaining on the cuticle of the adult, they separate from the adult cuticle and stick to the pupal case. As long 4s this occurs on parts of the wings only moths emerge normally but for the absence of scales on the wings. In the complete absence of wheat-germ oil the scales also remain stuck in other regions, especially over the abdomen, and thus emergence is made impossible. A very similar if not perhaps identical phenomenon has been described before as a gene mutation in the same species (Kühn & Henke, 1929), and we shall suggest below a possible explanation of this phenomenon.

In this paper we shall use terms such as scale deficiency, scale condition or scale factor when referring to a nutritional deficiency which affects the scales or emergence, as distinct from a growth deficiency whereby the rate of growth is affected. Both types of deficiency occur in Ephestia in connexion with fatsoluble substances in the diet. In E. kuehmella, growth is greatly retarded in the absence of wheatgerm oil, and increasing amounts of it have a similar improving effect on growth and scales. Fig. 2 shows the rate of growth of larvae, grown in the presence of varying amounts of wheat-germ oil, which have been analysed in Table 1 for the effect on scales and emergence. With 8·7 and 4·3 mg. of wheat-germ oil the rate of growth is optimal and the moths are normal. With decreasing amounts of wheat-germ oil rate of growth and condition of scales becomes increasingly worse, and both are at the minimum in the absence of wheat-germ oil. There was, however, an exception in the diet with 1·2 mg. wheat-germ oil where growth was very fast but the moth abnormal. In diets with 1·2 mg. and less the oil had been diluted with petroleum ether (tech., B.D.H.) and it was considered not impossible that the ether, or impurities in it, may have had a favourable effect on growth. Fast growth and good scales, therefore, do not necessarily always go together.

A third effect, arising from the absence of wheatgerm oil, is increased mortality of the larva (as distinct from mortality in the pupa at the time of emergence). The number of larvae ultimately pupating is always lower in the absence of wheat-germ oil, but differs widely on different occasions. In the tests of Fig. 2, eleven out of twenty pupated, in Figs. 9 and 5 the numbers were nine and three respectively.

E. elutella in the absence of wheat-germ oil never emerges from the pupa, and there are no signs of unsuccessful attempts at emergence (Table 2 (5)). Without the oil, growth is always delayed, but, as a rule, both growth rate and mortality are less affected than with E. kuehniella. No test has yet been performed with E. elutella with graded doses of wheatgerm oil, but from tests with graded doses of linoleic acid, to be reported below, there can be no doubt that essentially the same response would have been obtained as with E. kuehniella.

A third representative of the genus Ephestia, E. cautella, reacts very much in the same way in the absence of wheat-germ oil : growth is delayed and the moths fail to emerge from the pupal cases.

A fourth species investigated was Plodia inter - punctella which is the species most closely related to Ephestia. Plodia never failed to emerge in a perfect state in the absence of wheat-germ oil. Growth was retarded, but the extent of this delay was different on different occasions. It was at first thought that perhaps sufficient of the fat-soluble factors might have been supplied by the fat which is contained in 5 % yeast. Tests were consequently carried out with diets containing yeast which had been extracted with ether (Soxhlet). Here growth was slower than with normal yeast, but the moths always emerged normally.

Wheat-germ oil was subsequently saponified, and both the resulting fractions, saponifiable and unsaponifiable, were tested separately and combined (Fig. 3). With both fractions, growth is usually, but not always, rather slower than with wheat-germ oil, which indicates changes in the resulting fractions. With the saponifiable fraction alone, the moths of both kuehniella and elutella always emerge normally, but growth is usually (elutella) or always (kuehniella) slower than with both fractions. With kuehniella on certain occasions (Fig. 3), but not always (Figs. 9, 14), the saponifiable fraction did not accelerate growth. With elutella growth is relatively less impaired in the absence of the saponifiable fraction.

With the unsaponifiable fraction alone, the moths of both species fail to emerge, but growth is always considerably faster with it than Without it. The extent to which the growth rate is maintained again varies considerably on different occasions. In most experiments with E. kuehniella growth was slower with the saponifiable than with the unsaponifiable fraction alone, and fastest with both together (Fig. 3). On one occasion, however (Fig. 14) the saponifiable fraction was faster than the unsaponifiable fraction, but again growth was greatly superior in the presence of both. The growth-promoting effect of the unsaponifiable fraction was more marked with kuehniella than elutella. This is what one would have expected, as elutella grows relatively faster than kuehmella in the absence of wheat-germ oil. This difference in growth of these two species on a diet without wheat-germ oil is almost certainly due to the effect of yeast oil on elutella. On a fat-free diet, containing yeast which had been thoroughly extracted with ether (Soxhlet), E. elutella grew very much more slowly than with ordinary yeast, and growth was greatly improved by the addition of the unsaponifiable fraction of wheat-germ oil. It is, therefore, clear that both species require growth-promoting factors which may be supplied with the unsaponifiable fraction of wheat-germ oil.

E. cautella reacted very similarly to the other two species. With both fractions of wheat-germ oil present, normal moths emerged, and the moths were also normal with the saponifiable fraction alone. No moths emerged with the unsaponifiable fraction alone. Each fraction separately had a favourable effect on growth. Growth was delayed if the diet contained ether-extracted yeast.

It is therefore clear that for the three Ephestia species, emergence and good scales are bound up with a substance present in the saponifiable fraction of wheat-germ oil, while the unsaponifiable fraction contains something which improves growth.

Wheat-germ oil contains 50 –60 % of linoleic acid. A sample (from Vitamins Ltd.) which we used in the latter part of our experiments was analysed by Prof. Hilditch and found to contain 54 ·9 %. As linoleic acid is one of the essential fatty acids in the nutrition of the rat it was an obvious step to test its effect on Ephestia. In early experiments with a technical sample of linoleic acid (B.D.H.) normal moths emerged but growth remained slow. Later, pure samples were obtained by courtesy of the late Mrs Smedley-Maclean (methyl-linoleate) and Prof. G. O. Burr (ethyl-linoleate), and the results obtained with the technically impure sample were fully confirmed.

On E. kuehniella, addition of linoleic acid alone to the diet hardly improved the scale condition or growth, while with E. elutella, on occasions, perfectly formed moths were obtained. Tests with E. kuehniella on diets containing the unsaponifiable fraction of wheatgerm oil in addition to pure linoleic acid were at first unsuccessful. In those earlier experiments the unsaponifiable fraction had been prepared from a sample of wheat-germ oil kept for several years in the laboratory, which might have undergone changes. Subsequently, tests carried out with the unsaponifiable fraction of a fresh sample of wheat-germ oil were completely successful. The condition of the scales and emergence now became a function of the amounts of linoleic acid in the diet, exactly as had been the case in tests with graded doses of wheat-germ oil (Table 1 (2 a –e)). With 6 mg. of linoleic acid, perfectly formed moths emerged. With 4 mg. the first moths were also perfect specimens, but later four moths emerged with slightly defective wings. With the same quantity of linoleic acid, but without the unsaponifiable fraction, the condition of the wings was very much worse (Table 1 (3)). With 2 and 1 ·3 mg. of linoleic acid the wings became progressively mûre naked, and with 0 ·33 mg. the majority of the wings had no scales at all and many moths were badly formed. But even such a small quantity of linoleic acid had a very marked effect on emergence, as with no linoleic acid most moths failed to emerge and all that emerged were badly formed (Table 1 (7)).

While, therefore, linoleic acid was undoubtedly the factor necessary for perfect condition of the scales, it had in this experiment, paradoxically, an adverse effect on growth (Fig. 4). With increasing quantities, as the scale condition improved, the growth rate decreased, being fastest in the diet without linoleic acid when most moths failed to emerge, and slowest with the largest quantities of linoleic acid when the moths were normal. No such adverse effect of larger doses occurred in diets with graded doses of wheatgerm oil (Fig. 2). It is therefore obvious that the adverse effects must be attributed to a deterioration in the pure linoleic acid. This also follows from a comparison of the effects of identical amounts of linoleic acid in diets containing wheat-germ oil and diets with pure linoleic acid. As already stated above the wheatgerm oil which we used contained 55 % of linoleic acid. A perusal of the data in Table 1 makes it clear that identical quantities of linoleic acid were less efficient in the pure state than when supplied in wheat-germ oil. For instance, 2 ·4 mg. of linoleic acid contained in 4 ·3 mg. wheat-germ oil were sufficient, while even 4 mg. were insufficient when supplied in the pure state in the presence of the unsaponifiable fraction, and still more so when supplied alone (Table 1 (1 b, 2b and 3)). From tests to be described below there can be little doubt that the deterioration of linoleic acid is due to a rancidative oxidation which can be checked by adding vitamin E or other antioxidants.

Of other fatty acids tested, linolenic acid had a similar effect to linoleic acid. In the presence of the unsaponifiable fraction, and with 2 ·3 mg. linolenic acid the moths emerged perfectly and the growth rate was much better than with the unsaponifiable fraction alone (Fig. 5 and Table 1 (8)). This result was greatly superior to that obtained with a similar quantity of linoleic acid, both as regards scale condition and growth rate. But when comparing the effect with that of linoleic acid as a constituent of wheat-germ oil (Table 1) the result is about the same. Increasing the quantity of linolenic acid four times had an adverse effect on growth, obviously due to deterioration, but unexpectedly the scale condition also grew worse. Without the unsaponifiable fraction, linolenic acid deteriorated quickly. With 2 ·3 mg. only defective moths emerged, and none developed with the larger quantity (T able 1 (9)). In the presence of the saponifiable fraction of wheat-germ oil 2 ·3 mg. of lino¹eic acid did not improve growth and 9 ·2 mg. had a very detrimental effect.

We have so far not succeeded in procuring a good sample of arachidonic acid. An old discoloured specimen which had been kept for several years in an unsealed tube, and which was probably no longer in good condition, had no effect on scale condition. Oleic acid, of which a pure sample was provided by Dr G. King, had no effect whatsoever on scale condition but possibly a very slight positive effect on growth. Elaidic acid (also from Dr King) had no effect on scale condition, but affected the growth rate adversely. This latter may have been due to the high melting-point of this acid. The results of tests with the highly unsaturated C₂₀ and C₂₂ fatty acids from cod-liver oil will be described below.

With E. elutella a successful correction of the scale deficiency in the absence of the unsaponifiable factor was obtained at first with a technical sample of linoleic acid, and on one occasion only with pure linoleic acid. In other experiments linoleic acid alone never brought about emergence of the adult moth, but the scale-protecting effect of linoleic acid was fully operative in diets which contained in addition either the unsaponifiable fraction of wheat-germ oil, or vitamin E. A series of tests with graded doses of linoleic acid, and with the unsaponifiable fraction of wheat-germ oil, gave a graded response with regard to scale condition very similar to that obtained in the corresponding test with E. kuehniella (Table 2 (2)). Perfect moths emerged with 6 and 4 mg./g. of linoleic acid. With 2 and 1 ·3 mg. the wings became more and more defective. Few moths emerged with 0 ·33 mg. and none in the diet without linoleic acid. In the same test 4mg. linoleic acid, without the unsaponifiable fraction, had no effect whatsoever (Table 2 (3)). This test was carried out simultaneously with the corresponding test on E. kuehniella, and it is important to note that, unlike kuehniella, there was no obvious detrimental effect on growth in the diets with the higher quantities of linoleic acid. Speed of growth was rather variable, but showed no tendency to acceleration or delay according to the quantities of linoleic acid, and it was of the same order on diets without it.

A very similar result was obtained in tests with graded doses of linoleic acid and with vitamin E in place of the unsaponifiable fraction (Table 2 (6)). These tests were conducted at an earlier stage, before our classification of scale deficiences had been worked out, and early stages of deficiency may have been overlooked. It is almost certain that down to 3 ·3 mg. of linoleic acid the wings were normal. With 1 mg. they showed varying degrees of nakedness and hardly any moths emerged with 0 ·33 mg. (Table 2 (6e)). There was again no indication that linoleic acid had any effect, favourable or unfavourable, on growth. With 4 mg. of linoleic acid, and without vitamin E, none of the moths emerged, but unexpectedly growth was considerably faster than in all the other diets with linoleic and vitamin E. Some tests were also carried out with diets which contained ether extracted yeast. Here again, 5 mg. linoleic acid accelerated growth, and while normal moths appeared in the presence of vitamin E, none emerged in its absence (Fig. 6).

Wheat-germ oil contains relatively large amounts of unsaponifiable matter, most of which is probably sitosterol. The sample which we used in the latter parts of our experiments contained 8 % of unsaponifiable matter. It is now known that insects require sterols in the diet for growth (Fraenkel & Blewett, 1943), and to meet these requirements all the diets dealt with in this paper contained 1 % of cholesterol. There still remained a possibility that specific sterols in the unsaponifiable fraction of wheat-germ oil had an effect on the growth of Ephestia. Some tests were performed with several preparations of varying purity from wheat-germ oil (prepared by Glaxo Laboratories), and although the results were somewhat ambiguous it is almost certain that sterols play no part in the growth-promoting activity of the unsaponifiable fraction of wheat-germ oil.

From the results given in the foregoing section it is clear that the unsaponifiable fraction stabilizes linoleic acid. The unsaponifiable fraction of wheat-germ oil contains large amounts of vitamin E, and in view of the well-known function of vitamin E as an antioxidant, tests were performed with vitamin E in place of the unsaponifiable fraction. Two samples of tocopherols were used, dl-α-tocopherol from Glaxo Laboratories (‘ synthetic and probably well over 95 % pure’) and α-tocopherol acetate from Roche Ltd. These two samples did not always give exactly the same results, but it is not certain whether such differences were significant. When vitamin E was added to the saponifiable fraction of wheat-germ oil, the growth rate of E. kuehniella improved, but the extent of this improvement varied on different occasions. This may have been due to difficulties of dosage and to the instability of α-tocopherol. A series of tests with diets containing the saponifiable fraction of wheat-germ oil and graded doses of vitamin E, showed clearly an optimum effect with 0 ·31 mg. of vitamin E per g. of the diet (Fig. 7). With this concentration growth was greatly speeded up, much more so in this particular test than by the unsaponifiable fraction of wheat-germ oil. Larger doses (1 ·5 mg.) had a detrimental effect on growth, and with smaller doses (0 ·06 and 0 ·01 mg.) growth was also slower (Fig. 7). 0 ·33 mg./g. were subsequently used in most tests, but it must be remembered that with the narrow range of optimal efficiency and the well-known instability of α-tocopherol, optimal conditions cannot be expected to prevail in every experiment.

The amount of vitamin E in two samples of wheatgerm oil used by us was kindly estimated by Messrs Vitamins Ltd. and found to be 0 ·23 and 0 ·3 %, calculated as α-tocopherol. The amount present in a satisfactory diet, with wheat-germ oil, is therefore very much smaller than the amounts necessary when pure α-tocopherol is used. This is not surprising in view of the labile nature of this compound.

Subsequently, the effect of vitamin E was tested in combination with pure linoleic acid (Fig. 8). It will be remembered that with linoleic acid alone the moths either failed to emerge, or emerged with defective wings (Table 1 (3)). Perfectly formed moths emerged with 0 ·33 mg. vitamin E (Glaxo) and 4 mg. linoleic acid (Table 1 (5)), and growth was considerably better (Fig. 8) than in the corresponding test with the same quantity of linoleic acid and the unsaponifiable fraction of wheat-germ oil (Fig. 4). With vitamin E (Roche) the moths emerged with defective wings (Table 1 (4)), but growth was again better than with the unsaponifiable fraction of wheat-germ oil, where the moths had been perfect. Vitamin E, in this experiment, therefore, had as good or almost as good an effect on the scale condition as the unsaponifiable fraction of wheat-germ oil, and a better, effect on growth. This latter effect may perhaps be attributed to the larger amounts of vitamin E in the diets containing pure α-tocopherol. There can be therefore little doubt that most, if not all, of the favourable effect of the unsaponifiable fraction in the presence of the saponifiable fraction or linoleic acid is, due to its content of vitamin E.

Tests with E. elutella on the effect of vitamin E in combination with linoleic acid have already been described. The efficiency of vitamin E in stabilizing linoleic acid seems to be about the same as that of the unsaponifiable fraction of wheat-germ oil (Table 2).

The effect of vitamin E so far described raises the question of its specificity as an antioxidant. It was of interest to see whether other antioxidants had a similar effect in the nutrition of Ephestia. Tests with hydroquinone were unsuccessful, owing, possibly, to a toxic action. There was, however, a pronounced positive effect with gallates. Two preparations were used, ethylgallate (B.D.H.) and propylgallate (Nipa Laboratories), and the results obtained with both were similar. Whenever gallates were added to a diet which contained the saponifiable fraction of wheatgerm oil, there was a conspicuous improvement in the growth rate, but the acceleration was smaller than with the unsaponifiable fraction (Fig. 9). This improvement was most striking with large quantities of gallates (c.0 ·1 % of the diet, and 10 % of the fat) which, were incorporated in the diet in aqueous solution. With smaller quantities (0 ·01 % of the diet) the effect was better when the gallates were added in aqueous solution than when incorporated in the saponifiable fraction in alcoholic solution.

As a third type of antioxidant, ascorbic acid was added to a diet which contained the saponifiable fraction of wheat-germ oil. It was mixed into the diets in aqueous solution. With 0 ·33 mg. of vitamin C growth of E. kuehniella was slightly improved, while a ten times larger quantity seemed to have a detrimental effect (Fig. 10).

The question which remains to be answered is whether the antioxidant action of the unsaponifiable fraction, or vitamin E, on linoleic acid is the sole function of these substances in the nutrition of Ephestia. This does not seem very probable in view of the fact that in the absence of the saponifiable fraction the unsaponifiable fraction in most cases pronouncedly speeds up growth. Therefore the effect of adding vitamin E alone was investigated. Vitamin E on each of three occasions markedly improved growth of E. kuehniella to a very similar extent as the unsaponifiable fraction, and the moths, as one would have expected, failed to emerge from the pupae (Fig. 10). E. elutella, as stated above, sometimes grows relatively fast even in the absence of wheat-germ oil, and very much slower if the yeast in the diet has been extracted with ether. Vitamin E, added to a diet with extracted yeast, accelerated growth though not as much as did the ether extract of yeast (Fig. 6).

The effect of ethylgallate in the absence of the fat fraction of wheat-germ oil has so far been tested only once (1 mg./g.) and there was a slight improvement in the growth rate. The suggestions had been made above, in connexion with tests with graded doses of wheat-germ oil, that a technical sample of petrol ether, which had been used as a diluent, may have had an effect of its own on growth. Subsequently it was added to diets in similar amounts (1 ·3 –2 mg./g.), but in the absence of any oil or oil fraction, and on each of two occasions had a slightly favourable effect on growth (Fig. 10).

All the evidence so far available therefore points to the effect that certain antioxidants have a favourable influence on the growth rate of Ephestia, quite independent of the stabilizing effect which they have on dietary fats.

So far, we have dealt mainly with the effect of wheat-germ oil, or of fractions or substances derived from or contained in it. In the following section we shall describe the effects of fats from other sources.

In the summary of these tests, given in Table 3, it can be seen at a glance that the effect on the scale condition is very much dependent on the amount of linoleic acid which these fats contain. The effect on growth is very different with different fats as one would have expected, since the fats used probably differ very much in the degree of refinement, freshness and stability. Table 3 requires some explanations.

The sample used was prepared from fresh pig fat by mincing, boiling in water and straining through cloth. The linoleic acid content of our samples are not known. According to Hilditch, Lea & Pedelty (1939) linoleic acid in lard is derived entirely from the food, and typical examples contain 3 –8 %. The moths which emerged from diets containing lard invariably had partly naked wings, but the exact state was not investigated at the time (Table 1 (13)). The results obtained would be consistent with a linoleic acid content of the lard of c. 7 % (the effect of 10 mg. lard being approximately the same as that of 1·2 mg. of wheat-germ oil). While lard, therefore, proved a very moderate source for linoleic acid, it had a very pronounced effect on growth which in the best cases was as fast as on wheatgerm oil (Fig. 11). Lard contains c. 2 % of highly unsaturated C_20–22 acids which include arachidonic acid (Hilditch et al. 1939). From our results with cod-liver oil (see below) it is not impossible that these acids may improve growth. The saponifiable fraction of lard also improved growth, but to a lesser degree than whole lard. Lard is usually considered to be poor in natural antioxidants. It contains relatively little unsaponifiable matter, and we have not succeeded in obtaining sufficient for a test.

The effect; of lard on Ephestia closely resembles that on rats, where the skin condition is affected according to its linoleic acid content, while the effect on growth is also disproportionally superior (Hume, Nunn, Smedley-Maclean & Smith, 1938).

oil had a very pronounced growth effect (Figs. 12, 13) similar to lard, but the effect on the scales was entirely negative (Table 1 (10, 11)). Most moths failed to emerge, and those which did emerge were very badly formed. This fully agrees with what is known about the composition of these oils. According to Farmer & Van den Heuvel (1938a,b) cod-liver oil contains small amounts of polyethenoid acids with the following numbers of ethenoid linkings : C₁₆, 3 ; C₁₈, 4 ; C₂₀, 5 ; C₂₂, 6. Ordinary linoleic acid has not been detected. Halibut-liver oil contains no linoleic acid (Lovem, 1932). The fact that these two oils had no effect on the scale conditions made it the more surprising that there should have been such a good effect on growth. As this result very closely resembled that of similar work with rats (Burr, Burr & Brown, 1931 ; Burr, Brown, Kass & Lundberg, 1940; Hume et al. 1938) tests were carried out with different fractions from cod-liver oil, prepared by the method of fractional distilling and kindly put at our disposal by Dr E. H. Farmer. Of the various samples tested, a striking effect on growth was obtained with the pure methyl ester of docosahexenoic acid (Fig. 14). With 2 ·5 mg. of this acid, and in the presence of the unsaponifiable fraction of wheat-germ oil, there was a very considerable improvement in the growth rate if compared with the effect of the unsaponifiable fraction alone (which on this particular occasion had relatively little effect on growth). Four times this quantity of the C₂₂ acid had an inferior effect. Growth was not improved when this acid was supplied alone or together with the saponifiable fraction of wheatgerm oil, which clearly indicates that the C₂₂ acid is fiable to deteriorate unless stabilized by the unsaponifiable fraction. At the same time the C₂₂ acid had no effect whatsoever on scales or emergence. With the methyl ester of the unsaturated C₂₀ acid from cod-liver oil there was no similar effect on growth when applied in combination with the saponifiable or unsaponifiable fraction of wheat-germ oil, but there was a conspicuous growth effect when C₂₀ acids were added alone (Fig. 15). With mixtures of the saturated and unsaturated C₂₀ acids and mixtures of the saturated and unsaturated C₂₂ acids growth was much improved, and with a mixture of the C₂₂ plus higher methyl esters plus some unsaponifiable material slightly improved. These mixtures were tested in the absence of the saponifiable or unsaponifiable fraction of wheat-germ oil, and none of these mixtures had any effect on scales.

It seems therefore fair to conclude that the growth effect of cod-liver and possibly also halibut-liver oil is partly, if not entirely, due to its content of docosahexenoic and other unsaturated acids.

With 10 mg./g. of butter in the diet all the moths emerged from the pupae, but either badly formed, or with most of the scales missing on the wings (Table 1 (12 a)). With 2 ·5 mg. about half the moths failed to emerge (Table 1 (12b)). The effect of butter therefore seems to be somewhat inferior to that of lard, and consistent with a linoleic acid content of the order of 3 %. According to Hilditch (1940) milk fat contains only traces of ordinary linoleic acid, but ‘there is a certain proportion (3 –5 % of the total fatty acids) of octadecadienoic acids which appear to be different geometrical (cis-trans) isomerides of the Δ^9:10’^12:13-acid’. From the results obtained by us it therefore appears that this isomer is utilized by Ephestia to the same degree as linoleic acid from vegetable sources. Rats do not appear to utilize this acid (Burr, 1942).

Butter had a very marked positive growth effect, similar to that of cod-liver oil (Fig. 12). We do not know whether this is caused by this octodecadienoic acid, or antioxidants or something else. A superior growth of rats on diets with butter fat has been attributed to the presence, in butter, of small amounts of long-chain saturated fatty acids (Boutwell, Geyer, ElVehjem & Hart, 1941).

The sample used was obtained from dried brewers’ yeast by extraction with ether. There was no effect on the condition of the scales and emergence, neither with kuehniella, elutella or cautella, but there was a conspicuous improvement in growth in certain tests with elutella and cautella (but never with kuehniella) especially where ether extracted yeast was used in the diet (Fig. 6). Yeast oil contains a little linoleic acid (4 ·1 % according to Teufel, Thaler & Schreyegg, 1936), but it seems feasible that this might have been destroyed during the drying of the extracted oil at about 100° C. This treatment might also have affected unfavourably the antioxidant constituents and this is possibly the cause of the different effects on growth of different samples. Yeast oil is also inefficient in alleviating symptoms of a linoleic acid deficiency in the rat (Burr, 1942).

(from Prof. T. P. Hilditch). There was no effect on scale condition with the addition, of this oil alone, and a small effect on growth. With coconut oil, and in the presence of the unsaponifiable. fraction of wheat-germ oil, most moths of E. kuehniella and elutella emerged, but either incompletely formed or with the wings naked. The linoleic acid content of coconut oil is given as 1 –2 % (Hilditch, 1940), which is consistent with the results obtained on the diets with the unsaponifiable fraction of wheat-germ oil.

(from Prof. T. P. Hilditch) had an even smaller effect than coconut oil. In the presence of the unsaponifiable fraction of wheatgerm oil some very badly formed moths of E. kuehniella, but none of E. elutella, emerged, and growth in both species was slightly improved. With out the unsaponifiable fraction no moths emerged. Allenblackia oil is reported to contain only about 0 ·2 % of linoleic acid (Hilditch, 1940).

In our early experiments with E. kuehniella, long before the effects of wheat-germ oil were recognized, we had noticed that moths invariably emerged from the pupal case in the absence of wheat-germ oil if the diets contained starch in place of glucose. No close examination of the state of scales on the wings was made at the time, but it appears that with 50 % rice starch at least two out of nineteen emerged with bare wings, while the remaining seventeen, and also all the moths from a diet with 80 % starch, seem to have emerged with the wings either intact or not greatly deficient. A subsequent test (Fig. 16) with maize starch (80 %) yielded nine moths, five of which had defective wings. With wheat-germ oil added to this diet growth was considerably speeded up and all moths were normal. A simultaneous test with a sample of potato starch showed very bad growth. Only two pupae were formed and none emerged. With wheat-germ oil growth was still very slow but six perfect moths emerged. On a sample of potato starch which we prepared ourselves from fresh potatoes growth was relatively better, but again the moths only emerged in the presence of wheat-germ oil. In a third series of tests another sample of rice starch was used which had a peculiar flavour and may not have been in perfect condition. Here growth was very slow, both with and without wheat-germ oil, and without the oil altogether five moths emerged all with the scales missing.

It has been stated before that severe linoleic acid deficiency symptoms in rats do not develop on diets which contain starch (Burr, 1929; Evans, 1932), while deficiency symptoms readily appear if potato starch is used (Burr, 1942). According to Taylor & Lehrman (1926) cereal starches always contain some fat which is difficult to remove; the combined fatty matter in maize starch is about 0 ·5 %, and in rice starch 0 ·83%, with approximately 24% palmitic, 40 % oleic and 36 % linoleic acid. Lehrman & Kabat ( 1933) state that potato starch contains no combined fatty matter. The linoleic acid contents of the starches used by us are not known, but assuming them to be of the order of those investigated by Taylor & Lehrman, it would appear that our starch diets contained c. 0 ·2 % of linoleic acid, presumably in a relatively stable form. From what we know about requirements of linoleic acid by Ephestia, it seems feasible that these amounts may be sufficient to ensure the formation of some perfectly developed moths, and it does not seem necessary to postulate any special specific effect of starch, such as ‘refection ‘.

The investigation began with the discovery that wheat-germ oil was essential for the development of Ephestia in two ways. The caterpillars failed to grow or grew very slowly on artificial diets in the absence of wheat-germ oil, and those which ultimately pupated failed to emerge as adult moths. This twofold effect of wheat-germ oil on emergence and on growth could be localized in two different fractions : the saponifiable fraction was primarily necessary for emergence and good scales and the unsaponifiable fraction for growth. Further analysis led ultimately to the recognition of linoleic acid as the saponifiable and vitamin E as the unsaponifiable factor. The main complication in the interpretation of the results rests in the fact that these two factors not only show an independent action of their own on growth and scales, but also inter actions indispensable for the complete functioning of the mechanism. On the one hand, linoleic acid to have-an effect at all, must be stabilized by vitamin E or other antioxidants and, on the other hand, the growth promoting effect of vitamin E may be antagonized by the formation of toxic substances in the deterioration of linoleic acid, probably peroxides. These interactions made it difficult at first to recognize that linoleic acid has also a growth-promoting effect of its own.

There is now overwhelming evidence for the fact that linoleic acid constitutes the saponifiable factor necessary for good formation of scales and emergence. Graded doses of linoleic acid were shown to have an effect on scales exactly similar to that with graded doses of wheat-germ oil. With different oils the effect is closely parallel to their linoleic acid content. Further evidence is supplied by the positive effect of cereal starch, which, as is already known, usually contains small quantities of linoleic acid. Furthermore, linolenic acid has an identical or similar effect.

The exact physiological mechanism by which linoleic acid works is not yet known. All the known facts indicate that it is connected with the production or action of the moulting fluid. An insect during moulting or emergence from the pupal case excretes a fluid into the space between the old and new cuticle which facilitates the easy separation of the two. It is easy to see that the amount or the constitution of this fluid might influence emergence as shown by Ephestia. Without linoleic acid there is no emergence, with very little linoleic acid the cuticles separate, but the scales remain stuck to the wrong cuticle, and with more and more linoleic acid more and more scales remain in the right position. All we know about the development of the Lepidoptera wing in general, and of the wing of Ephestia especially, is compatible with this explanation (Köhler, 1932).

It is well known that linoleic acid easily deteriorates by rancid oxidation, with the formation of toxic peroxides (Burr & Barnes, 1943 a). This fully explains why linoleic acid has little or no effect, except in combination with the unsaponifiable fraction, vitamin E or other antioxidants. Perhaps the clearest example of this relationship is provided by the case where large quantities of linoleic acid produced perfect scales but bad growth, while smaller quantities produced bad scales and better growth. Obviously there had been some deterioration in the linoleic acid; with larger quantities more toxic substances were formed, while sufficient effective linoleic acid was retained. The antioxidant effect of vitamin E on linoleic acid has been mainly studied in connexion with oxidation of vitamin A by unsaturated fatty acids and its protection by vitamin E (Thompson & Steenbock, 1944; Dam, 1943, 1944; György & Tomarelli, 1944).

Considering that linoleic acid, in the absence of suitable antioxidants, easily deteriorates with the production of deleterious products, it is not surprising that growth in such cases is slow. This effect could easily be taken to indicate that the effect of linoleic acid was primarily concerned with scales, and not with growth and, moreover, would mask any active growth-promoting effect which linoleic acid might possess. There have been, in fact, several indications which make the assumption of such a growth promoting effect inevitable. On certain rare occasions linoleic acid, when added alone, or in combination with the unsaponifiable fraction or vitamin E, had a marked positive effect on growth, and such positive effects must have some significance. Again, linolenic acid, which obviously acts similarly to linoleic acid, had a pronounced growth effect, both when given alone and in combination with the unsaponifiable fraction. There have been many occasions when growth was faster in the presence of the saponifiable and unsaponifiable fractions than with unsaponifiable alone. Finally, it is conceivable that the growth promoting effect of the highly unsaturated C₂₀ and C₂₂ acids from cod-liver oil and possibly also from lard is essentially of the same nature, while it is impossible to attribute it to any antioxidant action. In fact, the pro-oxidant effect of these acids has been frequently noted in connexion with a destructive effect on vitamins A and E (Mattill & Golumbic, 1942; Dam, 1944).

Apart from the protective action on linoleic acid, vitamin E and probably other antioxidants also have a positive effect on growth in the absence of linoleic acid. This is most clearly shown by the effect of the unsaponifiable fraction alone on growth. Positive effects of a smaller order were also produced with pure a-tocopherols or gallates. In view of the labile nature of these substances and the toxic nature if given in excess, it is not surprising that the pure substances should have had a smaller effect than when given in the unsaponifiable fraction. Furthermore, the fact that in the presence of ample linoleic acid growth is sometimes very bad may be taken to indicate that vitamin E is being destroyed while exerting its antioxidant effect, an explanation given for the phenomenon that cod-liver oil may be the cause of a vitamin E deficiency (Mattill & Golumbic, 1942; Dam, 1944; Clausen, Bames & Burr, 1943). The idea that an antioxidant should have an effect of its own, independent of the presence of dietary fat, is not entirely new. The stability of adipose fat of rats, for instance, is dependent on the presence of antioxidants, such as vitamin E, which are entirely of dietary origin (Bames, Lundberg, Hanson & Burr, 1943). Insects contain relatively large quantities of fat in the fat body, and it is not difficult to imagine that the presence of antioxidants, originally derived from the food, may be of advantage expressed by a better growth rate. The effects of vitamin E deficiency in vertebrates are manifold, and it has been suggested that they may all be connected with the antioxidant function of vitamin E (Nutrition Reviews, 1945).

All our results can. therefore be explained by assuming that linoleic acid is necessary for good scales and good growth, while vitamin E is necessary for protecting linoleic acid, and, in addition, for good growth. In view of the highly labile nature of these substances it would have been surprising if the results obtained on different occasions had been exactly identical. Moreover, the conditions under which the tests were performed were highly unfavourable for work with such labile compounds. Very small amounts of these substances were evenly distributed in large amounts of casein and glucose. The diets were finely ground, air dried, and allowed a free circulation of air. They were kept unchanged for at least 4 –6 weeks in the case of favourable diets, and much longer with unfavourable ones. In these circumstances the conditions for oxidation, of Oxidizable substances are obviously very favourable. It would, of course, be preferable to change diets at frequent intervals. Unfortunately, such a course is impracticable, if not impossible, with Ephestia, as the larvae are hidden in a tangle of webbing and food and cannot easily be extracted without being damaged.

The effect of linoleic acid on rats, as shown by the work of Burr and Smedley-Maclean, is strikingly similar in many respects to the effect on our moths. The effect is mainly concerned with the skin; it is probably a coincidence that rats develop ‘scaly’ skins, while it is ‘scales’ on the wings which are affected in moths. Linoleic acid also shows a growth effect. Lard improves the skin condition according to the quantity of linoleic acid it contains, while there is a disproportionately superior effect on growth. The most remarkable agreement concerns the positive effect of cod-liver oil on growth, but not on scales, which appears to be to some extent due to the presence of docosahexenoic and other highly unsaturated acids. Linoleic acid deficiency does not develop fully in the presence of cereal starches, but it does with potato starch. The magnitude of requirements was difficult to determine in both cases. For rats it amounts to about of the diet (Burr & Bames, 1943b) while for Ephestia it seems to be ⁠. There are also certain discrepancies. Linolenic acid, according to most observers (Hume et al. 1938; Quackenbush, Kummerow & Steenbock, 1942), has a very inferior effect on growth and skin of rats, while for Ephestia it is at least equal to linoleic acid. Isolinoleic acid from butter has no effect on rats (Burr, 1942), while for Ephestia it seems to be as good as ordinary linoleic acid.

From the little known about the composition of insect fats (Hilditch, 1940) it is not possible to decide whether linoleic acid is a constant constituent. Very recently it was shown by Pepper & Hastings (1943) that the linoleic acid content of the fat of the sugar beet webworm, Loxostege sticticalis L., decreases during metamorphosis from 26% to less than 1 %, while oleic acid increases from 31 · 5 to 54 · 5%. This clearly indicates that linoleic acid may. be derived from the food and is not synthesized. We have so far failed to obtain any indication that linoleic acid is required by any insect other than the three Ephestia species. Even the closely allied species, Plodia interpunctella, which occurs together with Ephestia species on the same food does not require it. In the clothes moth, Tineolabisselliella, wheat-germ oil has a minute positive effect on growth (Fraenkel & Blewett, 1946 a). We have grown a number of beetle larvae belonging to the genera Tribolium, Tenebrio, Dermestes, Lasioderma, Sitodrepa and Ptinus in the complete absence of even traces of fat without any indication that fat, would have improved growth.

Finally, mention must be made of a genetical aberration in Ephestia kuehniella, described by Kilhn & Henke (1929) under the term Glasflügeligkeit, which appears to be morphologically indistinguishable from our nutritional Glasflügeligkeit. The variation series showing the degree of Glasflügeligkeit (Kühn & Henke, 1929, fig. 30, p. 76) is indistinguishable from our series obtained with graded doses of wheat-germ oil or linoleic acid, and in both cases the missing scales remain behind on the inside of the pupal case. This makes it probable that the physiological action of this particular gene is concerned with linoleic acid metabolism.

1. The larvae of the moths belonging to the genus Ephestia, E. kuehniella, E. elutella and E. cautella grow well on artificial diets which also contain wheat-germ oil. In the absence of wheat-germ oil growth is slow, the mortality high and the moths fail to emerge from the pupae.

2. With suboptimal quantities of wheat-germ oil moths emerge with the wings lacking in scales.

3. The saponifiable fraction of wheat-germ oil is necessary for emergence and good scales, and the unsaponifiable fraction for good growth.

4. The active substance in the saponifiable fraction is linoleic acid. Linolenic acid has approximately the same effect as linoleic acid. Oleic acid has no effect.

5. The active substance in the unsaponifiable fraction is vitamin E (α -tocopherol) and possibly other antioxidants. One of the functions of vitamin E is that of an antioxidant which stabilizes linoleic acid. In this respect vitamin E can be replaced by ethyl and propyl gallates, or by ascorbic acid.

6. Apart from this stabilizing effect on linoleic acid, the unsaponifiable fraction of wheat-germ oil, or vitamin E, have also an independent growth effect. There are also indications that linoleic acid has a growth effect.

7. It is shown that the effect of a number of vegetable and animal fats on scales and emergence is Strictly parallel to their linoleic acid content.

8. With cod-liver oil, halibut-liver oil, lard and butter growth is fast, while the effect on scales and emergence is small or nil. The growth effect of cod liver oil is mainly or partly due to its content of docosahexenoic and probably other highly unsaturated fatty acids.

9. Large quantities of cereal starches in the diet have a favourable effect on scales and emergence which is attributed to the presence, in starch, of impurities of linoleic acid. No such effect has been encountered with potato starch which is free of linoleic acid.

10. The similarities and dissimilarities in the linoleic acid deficiency of the rat and Ephestia have been discussed.

11. The symptoms of the linoleic acid deficiency of Ephestia are similar, if not perhaps identical, with those of a genetical condition known as Glasflügelig keit.

We gratefully acknowledge a grant by the Medical Research Council for the study of insect nutrition, of which this investigation is a part. We are also indebted to Prof. T. P. Hilditch (Liverpool) for analyses of wheat-germ oil and supplies of coconut and Allenblackia oil, the late Mrs I. Smedley-Maclean (Lister Institute) and Prof. G. O. Burr (Minneapolis) for gifts of linoleic acid, Dr E. H. Farmer (British Rubber Research Association) for gifts of linolenic acid and of docosahexenoic acid and other fatty acids isolated from cod-liver oil, Dr G. King (St Mary’s Jdospital) for gifts of oleic and elaidic acid, Vitamins Ltd. for a gift of wheat-germ oil and the analysis of its vitamin E content, and Glaxo Laboratories Ltd. for a gift of α -tocopherol.