The four possible combina­tions are: (1) violet—violet; (2) violet—white; (3) violet—white; (4) white—white. The first will result in pure violet flowers, the fourth in pure white, and the second and third in pale violet flowers. Since all four combina­tions will appear in equal numbers when the number of crossings is sufficiently large the numerical result will be:

violet : pale violet : white = 1 : 2 : 1

Fifty per cent. of the F₂ genera­tion will be pale violet, 25 per cent. violet, and 25 per cent. white. The violets and whites each will breed true when bred among themselves since they are pure, and produce only one type of eggs and pollen. The pale violets are hybrids and will again produce the two types of eggs and pollen, that is, if bred among themselves will again give violets, pale violets, and whites in the ratio 1:2:1. This the experi­ment confirms.

As has been stated, it not infrequently happens that all the hybrids of the first genera­tion are alike. In such cases the one character is “recessive,” i. e., overshadowed or covered by the other the “dominant” character, which alone appears in the hybrids. Thus when Mendel crossed peas having round seeds with peas having angular seeds all the hybrids had round seeds. The round form is dominant, the angular recessive, i. e., all the hybrids have round seeds. When these hybrids were bred among themselves the next genera­tion produced round and angular seeds in the ratio of 3:1 (5474 round to 1850 angular). The explana­tion is as follows. Let R denote round, A angular character; the pure breeds of parents have the gametic constitu­tion RR and AA respectively. When crossed, all the offsprings have the constitu­tion RA and since A is recessive this hybrid genera­tion resembles the pure RR parents. The F₁ genera­tion produces two kinds of eggs R and A and two kinds of pollen R and A in equal numbers, and these if inbred give the following four combina­tions in equal numbers:

RR, RA, AR, AA.

Since RA, AR, and RR all give round seeds the F₂ genera­tion produces round seeds to angular seeds in the ratio of 3:1. The two organisms with the gametic constitu­tion RR and RA look alike, yet they are different in regard to heredity. The gametically pure form RR is called homo­zygous, the impure form RA hetero­zygous.

2. W. S. Sutton203 was the first to show that the behaviour of the chromo­somes furnishes an adequate basis on which to account for Mendel’s law of the segrega­tion of the characters in the sex cells of the hybrids. If we disregard the cases of parthenogenesis and the X chromo­somes, we may state that each species is characterized by a definite number of chromo­somes, e. g.204

In the fertiliza­tion of the egg the number of chromo­somes is doubled (if we disregard for the moment the complica­tion caused by the X and Y chromo­somes which was considered in the previous chapter). It was noticed by Montgomery that each chromo­some had a definite size and individuality, and he suggested that homologous chromo­somes existed in sperm and egg and that in fertiliza­tion the homologous chromo­somes of egg and sperm always joined and fused in the special stage designated as synapsis, which will interest us later. On the basis of this sugges­tion Sutton developed the chromo­some theory of the mechanism of Mendelian heredity or segrega­tion.

According to this theory, all the cells of an individual (inclusive of the egg cells and sperm cells) have two sets of homologous chromo­somes, one from the father, the other from the mother. Before the egg and sperm are ready for the produc­tion of a new individual, each loses one set of homologous chromo­somes in the so-called reduc­tion division, but the lost set is made up indiscriminately of maternal as well as paternal chromo­somes, so that while one egg retains the maternal chromo­some A the other will retain the paternal one, and so on. If before the reduc­tion division all the eggs had the chromo­some constitu­tion AA₁, BB₁, CC₁, DD₁ (where A B C D are the paternal and A₁ B₁ C₁ D₁ the maternal chromo­somes), after the reduc­tion division each daughter cell has a full set of four chromo­somes, but maternal and paternal mixed. Thus the one cell may have AB₁CD₁, the other A₁B₁C₁D₁, etc. This, according to Sutton, is the basis of the Mendelian heredity. Suppose the determiner of a certain character (violet colour of flower in the bean) is located in a chromo­some A of this species. The homologous chromo­some in beans with white colour may be designated as a. According to the chromo­some theory of Mendelian heredity a differs from A in one point, though this difference is probably only of a chemical character and not visible.

If an egg with A is fertilized by a pollen with a (or vice versa), after fertiliza­tion the chromo­some constitu­tion of the fertilized egg is Aa. All the other homologous chromo­somes are identical and therefore need not be considered. All the nuclei of the F₁ genera­tion have the chromo­some constitu­tion Aa. All will form eggs and pollen with nuclei of the same chromo­some constitu­tion Aa, but all these sex cells will go through the matura­tion division before they are fertilized; and this reduc­tion division leads to the existence of two kinds of eggs in equal numbers, one containing only the A, the other only the a chromo­some; and the same happens in the pollen. When therefore the hybrids F₁ are mated among themselves, the following four chromo­some combina­tions will be produced:

Possible combina­tions in fertilized eggs AA, Aa, aa, in the ratio 1:2:1.

Now this is exactly the ratio of Mendelian heredity in the F₂ genera­tion. The plant with the chromo­some constitu­tion AA will form violet flowers, those with the chromo­some constitu­tion Aa will form pale violet flowers, and those with the chromo­some constitu­tion aa will form white flowers.

To quote Sutton’s words:

It is obvious that Sutton by this idea did for heredity in general what McClung had done for sex determina­tion or sex heredity, that is, he showed that the numerical results obtained in Mendelian heredity can be accounted for on the basis that factors for hereditary characters are carried by definite chromo­somes. The cytological basis of sex determina­tion becomes only a special case of the cytological basis of Mendelian heredity. In the examples quoted the plants giving rise to violet and to white flowers are homo­zygous for the colour of flower having the chromo­some constitu­tion AA and aa respectively; while the plants with pale violet flowers are hetero­zygous, having the chromo­some constitu­tion Aa in their nuclei. The former give rise to identical sex cells A and A or a and a; while the hetero­zygous plants give rise to different sex cells A and a.

From this point of view in Drosophila (and very probably also in man) the female is homo­zygous for sex having in all its cells the critical chromo­some constitu­tion XX and giving rise to one type of eggs only, each with one X chromo­some; while the male in these forms is hetero­zygous for sex having in all its cells the chromo­some constitu­tion XY and forming two different types of spermatozoa in equal numbers X and Y. In Abraxas and in the fowl the female is hetero­zygous for sex and the male homo­zygous.

3. If the chromo­somes are the vehicle for Mendelian heredity it should be possible to show that the various hereditary characters which follow Mendel’s law must be distributed over the various chromo­somes; and it should be possible to find out which characters are contained in the same chromo­some. It has already been stated that sex-linked heredity is intelligible on the assump­tion that the X chromo­some carries the sex-linked characters. T. H. Morgan and his pupils have shown with the greatest degree of probability that corresponding linkages occur in the other chromo­somes and that there are in Drosophila exactly as many groups of linkage as there are different chromo­somes, namely four.205

Mendel had found that when he crossed two species of peas differing in regard to two pairs of characters, he obtained in the F₂ genera­tion results which he calculated on the assump­tion that the segrega­tion of the two pairs of characters in the sex cells of the hybrids took place independently of each other. To illustrate by an example: When crossing a yellow round pea with a green wrinkled variety in which the characters round and yellow are dominant, green and wrinkled recessive, all the hybrids of the F₁ genera­tion had the characters round and yellow. When these were inbred the F₂ genera­tion produced four types of seed in the ratio 9: 3: 3: 1, namely:

(1) yellow round    (315 seeds)
(2) yellow wrinkled (101 seeds)
(3) green round    (108 seeds)
(4) green wrinkled  (32 seeds)

The explana­tion according to Mendel’s theory is as follows: Since the segrega­tion of each pair of characters occurs independently, there must be 3 yellow to 1 green and also 3 round to 1 wrinkled in the F₂ genera­tion. The yellow will, therefore, be round and wrinkled in the ratio of 3:1, which will give 9 yellow round to 3 yellow wrinkled. The green will also be round and wrinkled in the ratio of 3:1, which will give 3 green round to 1 green wrinkled, which is the ratio of 9: 3: 3: 1 found by Mendel.

On the basis of the chromo­some theory the following explana­tion could be given of this numerical rela­tion. The peas with yellow round seeds have sex cells with a factor for both yellow and for round in two different chromo­somes; these two different chromo­somes we will designate with Y and R. The peas with green and wrinkled seeds will have in their sex cells factors for these characters in two homologous chromo­somes g and w, where g is the homologue of Y and w of R. The cells of the hybrids of the F₁ genera­tion will have the chromo­some constitu­tion Yg Rw, where Y and g and R and w are homologous chromo­somes which will lie alongside each other YRgw. In the forma­tion of sex cells a reduc­tion of these four chromo­somes to two takes place whereby, according to the theory of Sutton, the following two types of separa­tion can take place: YR and gw, or gR and Yw. (A separa­tion into Yg and Rw is impossible since the division takes place only between homologous chromo­somes.) Hence there will be four types of eggs, YR, gw, gR, and Yw and the same four types of pollen cells. The F₂ genera­tion will produce the sixteen possible combina­tions in equal numbers: namely,

YRYR  YRgw  YRgR  YRYw
gwYR  gwgw  gwgR  gwYw
gRYR  gRgw  gRgR  gRYw
YwYR  Ywgw  YwgR  YwYw

Since w and g are recessives and therefore disappear when in combina­tion with their respective dominants Y and R the result will be 9 YR (yellow round), 3 Yw (yellow wrinkled), 3 Rg (round green), and 1 gw (green wrinkled) as Mendel actually observed and as all investigators since have confirmed.

Bateson made the discovery that these Mendelian ratios 9: 3: 3: 1 did not always occur when forms differing in two characters were crossed. He found typical and very constant devia­tions from this ratio in definite cases and these cases he interpreted as being due to “gametic coupling.”

Morgan has given an ingenious explana­tion of these devia­tions on the basis of the chromo­some theory of Mendelian heredity. He assumes that they occur in those cases where the two or more characters are contained in the same chromo­some. In that case the two factors lying in the same chromo­some should generally be found together. Such was the case for instance in the experi­ments with flies having red eyes and yellow body colour versus white eyes and grey body colour, the character for white eyes and yellow body being located in the X chromo­some (see preceding chapter), or in the experi­ments on Abraxas. These phenomena are called linkage, and the numerical results of linkage were given in the preceding chapter in connec­tion with the crossing of sex-linked characters.

We have already mentioned that before the matura­tion division occurs the homologous maternal and paternal chromo­somes fuse—the so-called synapsis of the cytologists—and afterward separate again. It had been observed by Janssens that in this stage of fusion and subsequent separa­tion a partial twisting and a partial exchange between two chromo­somes may take place. Morgan assumes that this exchange accounts for certain devia­tions in the ratio of linkage. If in Fig. 40 the white and black signify two homologous chromo­somes I and I₁ containing the two pairs of homologous factors AB and ab respectively, the synapsis state would be as in Fig. 41. If the separa­tion were complete, either I or its homologue I₁ might be lost in the matura­tion division of the egg. If, however, the synapsis is slightly irregular, as in Fig. 42, where the chromo­somes are slightly twisted, I and I₁ will not separate completely but an exchange will take place, part of I₁ and I becoming exchanged. This would result in the forma­tion of two mixed chromo­somes Ab and aB (Fig. 42). This partial exchange of homologous chromo­somes, which Morgan calls “crossing over,” occurs, as he found in Drosophila, in the egg only, not in the matura­tion division of the sperm. He informs me that in the silkworm moth Tanaka found that it occurs only in the male, while in Primula it takes place both in the ovules and in the pollen as shown by Gregory.

Morgan and his fellow-workers have put this theory to numerous tests by breeding experi­ments and the results have fully supported it. According to the chromo­some theory linkage should occur only when factors lie in the same chromo­some. Hence it should be possible, on the basis of this linkage theory, to foretell how many linkage groups there may occur in a species; namely, as many as there are chromo­somes. In Drosophila there are four pairs of chromo­somes, and Morgan and his fellow-workers found only four groups of linked characters.207 This agreement can be no mere accident.

Carrying the assump­tion still farther, these authors were able to show that each individual character has in all probability a definite loca­tion in the chromo­some, so that it seems as if each individual chromo­some consisted of a series of smaller chromo­somes, each of which may be a factor in the determina­tion of a hereditary character which is transmitted according to Mendel’s law of segrega­tion. Biology has thus reached in the chromo­some theory of Mendelian heredity an atomistic concep­tion, according to which independent material determiners for hereditary characters exist in a linear arrangement in the chromo­somes.

II

4. We are not concerned in this volume with the many applica­tions of the theory of heredity to the breeding of plants, animals, and man; the reader will find a discussion of these topics in the numerous writings of the special workers on genetics.208 We are, however, interested in the bearing this work has on the concep­tion of the organism. Two ques­tions present themselves: Is the organism nothing but a mosaic of hereditary characters determined essentially by definite elements located in the chromo­somes; and if this be true, what makes a harmonious whole organism out of this kaleidoscopic assortment? We call it a kaleidoscopic assortment since a glance at the list of hereditary characters found in one chromo­some, according to Morgan, shows that there is apparently no physio­logical or chemical connec­tion between them, and second: How can a factor contained in the chromo­some determine a hereditary character of the organism? To the first ques­tion we venture to offer the answer which has been already suggested in various chapters of this book, that the cytoplasm of the egg is the future embryo in the rough; and that the factors of heredity in the sperm only act by impressing the details upon the rough block. This metaphor will receive a more definite meaning by the answer to the second ques­tion. The characters which follow Mendelian heredity are morpho­logical features as well as instincts. For the former we have already had occasion to show in previous chapters to what extent they depend upon the internal secre­tions or the existence of specific compounds in the circula­tion, and the same is true for the instincts (Chapters VIII and X). This then leads us to the sugges­tion that these determiners contained in the chromo­somes give rise each to the forma­tion of one or more specific substances which influence various parts of the body. We probably do not notice all the effects in each case, but when a special organ is affected in a conspicuous way, we connect the factor with this organ or the special feature of the organ which is altered, and speak of a determiner or factor for that organ, or for one of its characters. We also understand in this way why outside condi­tions should be able to overcome the hereditary tendency in certain cases, for instance why the influence of certain hereditary factors for pigmenta­tion should depend upon temperature as E. Baur observed.

The view, according to which the determiners in the chromo­somes only tend to give special characters to the embryo or to the adult while the cytoplasm of the egg may be considered the real embryo, receives some support from the fact that the first development of the egg is purely maternal, even if the egg nucleus has been replaced by sperm of a different species. If an egg of a sea urchin be cut into two pieces, one with and one without a nucleus, and the enucleated piece be fertilized with the sperm of a different species of sea urchin, the blastula and gastrula stages are purely maternal and only the skeleton of the pluteus stage begins to betray the influence of the foreign sperm inasmuch as this skeleton is purely paternal, according to Boveri. In all experi­ments on hybridiza­tion it has been found that the rate of cell division of the egg is a purely maternal character. Thus when fish eggs of a species, in which the rate of first segmenta­tion of the egg is about eight hours, are fertilized with sperm of a species for which the same process requires about thirty minutes or less at the same temperature, the rate of segmenta­tion is again about eight hours. There is then no chromo­some influence noticeable in the early development.

When two forms of sea urchins, Strongylo­centrotus franciscanus and purpuratus,209 are crossed, certain features of the skeleton of the embryo, e. g., the so-called cross-bars, are a dominant, inasmuch as they are found in purpuratus and both the crosses, while they are absent in franciscanus. The development prior to the forma­tion of the skeleton is purely maternal. These observa­tions again lend support to the idea that the Mendelian factors of heredity must have the embryo to work on and that the organism is not to be considered a mere mosaic of Mendelian factors. This is further supported by the idea that the species specificity resides in the proteins of the unfertilized egg (see Chapter III), and it is quite likely that this species specificity decides which type of animal should arise from an egg.

The idea had been suggested that the factors which determine the future character might be ferments or enzymes, or substances from which such ferments develop. A. R. Moore210 pointed out that the cross-bars in the skeleton of the hybrid between S. purpuratus and franciscanus develop more slowly than in the pure breed and that this should be expected if the determiners were enzymes. Since the pure purpuratus has two determiners for the development of the cross-bars (from both egg and sperm), the hybrids only one (from either egg or sperm), the pure purpuratus should have twice the enzyme mass of the hybrid. It is known that the velocity of a chemical reac­tion increases in propor­tion with the mass (or in some cases in propor­tion with the square root of the mass) of the enzyme; the cross-bars should therefore develop faster in the pure than in the hybrid breeds, as was observed by Moore. It was, however, not possible to obtain quantitative data.

On the other hand, it is obvious that this reasoning would not hold for all cases. Thus when beans with violet flowers are crossed with white-flowered beans the hybrids are pale blue, which indicates that the hybrids have less pigment than the pure violet. Now we know that the mass of enzyme does not influence the chemical equilibrium but only the velocity of the reac­tion. The hybrids and pure violets differ, however, in the mass of violet pigment formed, that is to say, in regard to the equilibrium. Hence the idea that the determiners are enzymes or give rise to enzymes is probably not applicable to cases of this type.

The experi­ments on the heredity of pigments are at present almost the only ones which can be used for an analysis of the chemical nature of the character and its possible determiner. The important work of G. Bertrand211 and of Chodat212 on the produc­tion of black pigment in the cells of animals and plants with the aid of enzymes has paved the way for such work. Bertrand has shown that tyrosine (p-oxy­phenyl­amino­propionic acid) is trans­formed into a black pigment by an enzyme tyrosinase which occurs in numerous organisms and is obviously the cause of pigment and coloura­tion in a great number of species. This discovery was utilized in the study of the heredity of pigments by Miss Durham, Gortner,213 and very recently by Onslow.214 The latter showed that from the skins of certain coloured rabbits and mice a peroxidase can be extracted which behaves like a tyrosinase toward tyrosine in the presence of hydrogen peroxide. This peroxidase was found in the skins of black agouti, chocolate and blue rabbits, but not in yellow or orange rabbits. The recessive whiteness in rabbits and mice according to this author is due to the lack of the peroxydase. There exists a dominant whiteness in the English rabbit which is due to a tyrosinase inhibitor which destroys the activity of the tyrosinase “and the dominant white bellies of yellow and agouti rabbits are due to the same cause.” “Varia­tions in coat colour are probably due to a quantitative rather than to a qualitative difference in the pigment present.”

One point might still be mentioned since it may help to overcome a difficulty in visualizing the connec­tion between the localiza­tion of a factor in the chromo­some and the produc­tion of a comparatively large quantity of a specific chemical compound, e. g., a chromogen or a tyrosinase. We must remember that all the cells of an organism have identical chromo­somes, so that a factor for an enzyme like tyrosinase is contained in every cell throughout the whole body. It is likely, however, that the same factor (which we may conceive to be a definite chemical compound) will find a different chemical substrate to work on in the cells of different organs of the body, since the different organs differ in their chemical composi­tion. Thus it is conceivable that in the produc­tion of tyrosinase or of tyrosine not a single chromomere of one single cell is engaged, but the sum total of all these individual chromomeres of all the cells in one or several organs of the body. The writer has added this remark especially in considera­tion of the fact that some authors seem to feel that the chromo­some concep­tion of heredity is incompatible with a physico­chemical view of this process.

Since we have mentioned this difficulty which some writers seem to find in the chromo­some theory of Mendelian heredity, it may be added that a single factor may suffice to determine a series of complicated reflexes. Thus the helio­tropic reac­tions of animals are due to the presence of photo­sensitive substances, and it suffices for the hereditary transmission of the complicated purposeful reac­tions based on these tropisms that a factor for the forma­tion of the photo­sensitive substance should exist.215

5. Another point should be emphasized, namely that for Mendelian heredity it is immaterial whether the character is introduced by the spermato­zoön or by the egg. This fact which Mendel himself already recognized is in full harmony with the conclusion that the chromo­somes and not the cytoplasm are the bearers of Mendelian heredity, since only in respect to the chromo­some constitu­tion are egg and sperm alike, while they differ enormously in regard to the mass of protoplasm they carry. We can, therefore, be tolerably sure that wherever we deal with a hereditary factor which is determined by the egg alone the cytoplasm of the latter is partly or exclusively responsible for the result.

We have already mentioned the fact that the rate of segmenta­tion of the egg is such a character. Yet this character is as definite as any Mendelian character, and it would be as easy to discriminate two species of eggs by the time required from insemina­tion to the beginning of cell division as it would be by any Mendelian character of their parents.

The applica­tion of our modern knowledge of heredity to human affairs has been discussed in a very original way by Bateson in his address before the British Associa­tion in Sydney to which the reader may be referred.216

CHAPTER X

ANIMAL INSTINCTS AND TROPISMS217

1. The idea that the organism as a whole cannot be explained from a physico­chemical viewpoint rests most strongly on the existence of animal instincts and will. Many of the instinctive actions are “purposeful,” i. e., assisting to preserve the individual and the race. This again suggests “design” and a designing “force,” which we do not find in the realm of physics. We must remember, however, that there was a time when the same “purposefulness” was believed to exist in the cosmos where everything seemed to turn literally and metaphorically around the earth, the abode of man. In the latter case, the anthropo- or geocentric view came to an end when it was shown that the motions of the planets were regulated by Newton’s law and that there was no room left for the activities of a guiding power. Likewise, in the realm of instincts when it can be shown that these instincts may be reduced to elementary physico­chemical laws the assump­tion of design becomes superfluous.

If we look at the animal instincts purely as observers we might well get the impression that they cannot be explained in mechanistic terms. We need only consider what mysticism apparently surrounds all those instincts by which the two sexes are brought together and by which the entrance of the spermato­zoön into the egg is secured; or the remarkable instincts which result in providing food and shelter for the young genera­tion.

We have already had occasion to record some cases of instincts which suggest the possibility of physico­chemical explana­tion; for example the curious experi­ment of Steinach on the reversal of the sexual instincts of the male whose testes had been exchanged for ovaries. There is little doubt that in this case the sexual activities of each sex are determined by specific substances formed in the interstitial tissue of the ovary and testes. The chemical isola­tion of the active substances and an investiga­tion of their action upon the various parts of the body would seem to promise further progress along this line.

Marchal’s observa­tions on the laying of eggs by the naturally sterile worker wasps are a similar case. The fact that such workers lay eggs when the queen is removed or when they are taken away from the larvæ may be considered as a manifesta­tion of one of those wonderful instincts which form the delight of readers of Maeterlinck’s romances from insect life. Imagine the social foresight of the sterile workers who when the occasion demands it “raise” eggs to preserve the stock from extinc­tion! And yet what really happens is that these workers, when there are no larvæ, can consume the food which would otherwise have been devoured by the larvæ; and some substance contained in this food induces the development of eggs in the otherwise dormant ovaries. What appeared at first sight as a mysterious social instinct is revealed as an effect comparable to that of thyroid substance upon the growth of the legs of tadpoles in Gudernatsch’s experi­ment (Chapter VII).

2. If we wish to show in an unmistakable way the mechanistic character of instincts we must be able to reduce them to laws which are also valid in physics. That instinct, or rather that group of instincts, for which this has been accomplished are the reac­tions of organisms to light. The reader is familiar with the tendency of many insects to fly into the flame. It can be shown that many species of animals, from the lowest forms up to the fishes, are at certain stages—very often the larval stage—of their existence, slaves of the light. When such animals, e. g., the larvæ of the barnacle or certain winged plant lice or the caterpillars of certain butterflies, are put into a trough or test-tube illuminated from one side only, they will rush to the side from which the light comes and will continue to do this whenever the orienta­tion of the trough or test-tube to the light is changed; while they will be held at the window side of the vessel if the light or the posi­tion of the vessel remains unchanged. This instinct to get to the source of light is so strong that, e. g., the caterpillars of Porthesia chrysorrhœa die of starva­tion on the window side of the vessel, with plenty of food close behind. This powerful “instinct” is, as we intend to show, in the last analysis, the expression of the Bunsen-Roscoe law of photo­chemical reac­tions. A large number of chemical reac­tions are induced or accelerated by light, and the Bunsen-Roscoe law shows that the chemical effect is in these cases, within certain limits, equal to the product of the intensity into the dura­tion of illumina­tion.

The “attraction” or “repulsion” of animals by the light had been explained by the biologists in an anthropo­morphic way by ascribing to the animals a “fondness” for light or for darkness. Thus Graber, who had made the most extensive experi­ments, gave as a result the statement that animals which are fond of light are also fond of blue while they hate the red, and those which are fond of the “dark” are fond of red and hate the blue.218 In 1888 the writer published a paper in which he pointed out that the so-called fondness of animals for light and blue and for dark and red was simply a case of an automatic orienta­tion of animals by the light comparable to the turning of the tips of a plant towards the window of the room in which the plant is raised.219

The phenomenon of a plant bending or growing to the source of light is called positive helio­tropism (while we speak of negative helio­tropism in all cases in which the plant turns away from the light, as is observed in many roots). The writer pointed out that animals which go to the light are positively helio­tropic (or photo­tropic) and do so because they are compelled automatically by the light to move in this direc­tion, while he called those animals which move away from the light negatively helio­tropic; they are automatically compelled by the light to move away from it. What the light does is to direct the motions of the animals and to explain this the following theory was proposed. Animals possess photo­sensitive elements on the surface of their bodies, in the eyes, or occasionally also in epithelial cells of their skin. These photo­sensitive elements are arranged symmetrically in the body and through nerves are connected with symmetrical groups of muscles. The light causes chemical changes in the eyes (or the photo­sensitive elements of the skin). The mass of photo­chemical reac­tion products formed in the retina (or its homologues) influences the central nervous system and through this the tension or energy produc­tion of the muscles. If the rate of photo­chemical reac­tion is equal in both eyes this effect on the symmetrical muscles is equal, and the muscles of both sides of the body work with equal energy; as a consequence the animal will not be deviated from the direc­tion in which it was moving. This happens when the axis or plane of symmetry of the animal goes through the source of light, provided only one source of light be present. If, however, the light falls sidewise upon the animal, the rate of photo­chemical reac­tion will be unequal in both eyes and the rate at which the symmetrical muscles of both sides of the body work will no longer be equal; as a consequence the direc­tion in which the animal moves will change. This change will take place in one of two ways, according as the animal is either positively or negatively helio­tropic; in the positively helio­tropic animal the resulting motion will be toward, in the negatively helio­tropic from, the light. Where we have no central nervous system, as in plants or lower animals, the tension of the contractile or turgid organs is influenced in a different way, which we need not discuss here.

The reader will perceive that according to the writer’s theory two agencies are to be considered in these reac­tions: first, the symmetrical arrangement of the photo­sensitive and the contractile organs, and second, the relative masses of the photo­chemical reac­tion products produced in both retinæ or photo­sensitive organs at the same time. If a positively helio­tropic animal is struck by light from one side, the effect on tension or energy produc­tion of muscles connected with this eye will be such that an automatic turning of the head and the whole animal towards the source of light takes place; as soon as both eyes are illuminated equally the photo­chemical reac­tion velocity will be the same in both eyes, the symmetrical muscles of the body will work equally, and the animal will continue to move in this direc­tion. In the case of the negatively helio­tropic animal the picture is the same except that if only one eye is illuminated the muscles connected with this eye will work less energetically. The theory can be nicely tested for negatively helio­tropic animals in the larvæ of the blowfly when they are fully grown, and for positively helio­tropic animals on the larvæ of Balanus, and many other organisms.

One of the difficulties in identifying the motions of animals to or from the light with the positive and negative helio­tropism of plants consisted in the fact that plants are mostly sessile (and respond to a one-sided illumina­tion with helio­tropic curvatures to or from the light), while most animals are free moving and respond to the one-sided illumina­tion by being turned and compelled to move to or from the light. This difficulty was overcome by the observa­tion that sessile animals like the hydroid Eudendrium (Fig. 43) or the tube worm Spirographis (Fig. 44) react to a one-sided illumina­tion also with helio­tropic curvatures like sessile plants.220 On the other hand, it had been found before by Strassburger that free-swimming plant organisms like the swarmspores of algæ move to or from the source of light as do free-swimming animals.

3. The writer suggested in 1897221 that the light acts chemically in the helio­tropic reac­tions and in 1912 that the helio­tropic reac­tions probably follow the law of Bunsen and Roscoe,222 and it was possible to confirm this idea by direct experi­ments.223 This law states that the photo­chemical effect of light equals i t where i is the intensity of the light and t the dura­tion of illumina­tion. The experi­ments were carried out on young regenerating polyps of Eudendrium by measuring the time required to cause fifty per cent. of the polyps to bend to the source of light. The intensity of light was varied by altering the distance of the source of light from the polyps. Table VI gives the result.

TABLE VI

We must therefore conclude that the helio­tropic curvature of the polyps is determined by a photo­chemical action of the light. The light brings about or accelerates a chemical reac­tion which follows the Bunsen-Roscoe law. As soon as the product of this reac­tion on one side of the polyp exceeds that on the other by a certain quantity, the bending occurs. When the product i t is the same for symmetrical spots of the organism no bending can result. This is what our theory suggested.

It is very difficult to prove directly the applicability of the Bunsen-Roscoe law for free-moving animals, but it can be shown that intermittent light is as effective as constant light of the same intensity, provided that the total dura­tion of the illumina­tion by the intermittent light is equal to that of the constant light, and the dura­tion of the intermission is sufficiently small (Talbot’s law). Talbot’s law is in reality only a modifica­tion of the Bunsen-Roscoe law. Ewald has proved in a very elegant way the applicability of Talbot’s law to the orienta­tion of the eyestalk of Daphnia.224 This makes it probable that the law of Bunsen-Roscoe underlies generally the helio­tropic reac­tion of animals.

It is of importance for the theory of the identity of the helio­tropism of animals and plants that in the latter organisms the law of Bunsen and Roscoe is also applicable. This had been shown previously by Fröschel225 and by Blaauw.226 In the following table are given the results of Blaauw’s experi­ments on the applicability of the Bunsen-Roscoe law for the helio­tropic curvature of the seedlings of oats (Avena sativa). The time required to cause helio­tropic curvatures for intensities of light varying from 0.00017 to 26520 metre-candles was measured. The product i t, namely metre-candles-seconds, varies very little (between 16 and 26).

TABLE VII

It is, therefore, obvious that the blind instinct which forces animals to go to the light, e. g., in the case of the moth, is identical with the instinct which makes a plant bend to the light and is a special case of the same law of Bunsen and Roscoe which also explains the photo­chemical effects in inanimate nature; or in other words, the will or tendency of an animal to move towards the light can be expressed in terms of the Bunsen-Roscoe law of photo­chemical reac­tions.

The writer had shown in his early publications on light effects that aside from the helio­tropic reac­tion of animals, which as we now know depends upon the product of the intensity and dura­tion of illumina­tion, there is a second reac­tion which depends upon the sudden changes in the intensity of illumina­tion. These latter therefore obey a law of the form: Effect = f (di/dt).227 Jennings has maintained that the helio­tropic reac­tions of unicellular organisms are all of this kind, but investiga­tions by Torrey and by Bancroft228 on Euglena have shown that Jennings’s statements were based on incomplete observa­tions.

4. In these experiments only one source of light was applied. “When two sources of light of equal intensity and distance act simultaneously upon a helio­tropic animal, the latter puts its median plane at right angles to the line connecting the two sources of light.”229 This fact has been amply verified by Bohn, by Parker and his pupils, and especially by Bradley Patten, who used it to compare the relative efficiency of two different lights.

The behaviour of the animals under the influence of two lights is a confirma­tion of our theory of helio­tropism inasmuch as the animal moves in such a direc­tion that the symmetrical elements of the surface of the body are struck by light of the same intensity at the same angle, so that as a consequence equal masses of photo­sensitive substances are produced in symmetrical elements of their eyes or skin in equal times. The effect on the symmetrical muscles will be identical. As soon as one of the lights is a little stronger the animal will deviate towards this light, in case it is positively helio­tropic and towards the weaker light if it is negatively helio­tropic. This devia­tion again is not the product of chance but follows a definite law as Patten230 has recently shown. He used the negatively helio­tropic larvæ of the blowfly. These larvæ were made to record their trail while moving under the influence of the two lights. The results of the measurements of 2500 trails showing the progressive increase in angular devia­tion of the larvæ (from the perpendicular upon the line connecting the two lights), with increasing differences between the lights, are given in the following table. Since the devia­tion or angular deflec­tion of the larvæ is towards the weaker of the two lights it is marked negative.

TABLE VIII

Let us assume that the negatively helio­tropic animal is at an equal distance from the two unequal lights and placed so that at the beginning of the experi­ment its median plane is at right angles to the line connecting the two lights, but with its head turned away from them. In that case the velocity of reac­tion in the symmetrical photo­sensitive elements of the eyeless larvæ is greater on the side of the stronger light. Since the animal is negatively helio­tropic this will result in a greater relaxa­tion or a diminu­tion of the energy produc­tion of the muscles turning the head of the animal towards the side of the stronger light. Hence the animal will automatically deviate from the straight line towards the side of the weaker light. By the altera­tion of the posi­tion of its body the photo­sensitive elements exposed to the stronger of the two lights will be put at a less efficient angle and hence the rate of photo­chemical reac­tion on this side will be diminished. The devia­tion from the perpendicular in which the animal will ultimately move will be such that as a consequence, the rate of photo­chemical reac­tion in symmetrical elements is again equal. The ultimate direc­tion of motion will, according to our theory always be such that the mass of chemical products formed under the influence of light in symmetrical photo­sensitive elements during the same time is equal.

Patten also investigated the ques­tion whether the same difference of percentage between two lights would give the same devia­tion, regardless of the absolute intensities of the lights used. The absolute intensity was varied by using in turn from one to five glowers. The relative intensity between the two lights varied in succession by 0, 8¹⁄₃, 16²⁄₃, 25, 33¹⁄₃, and 50 per cent. Yet the angular deflec­tions were within the limits of error identical for each relative difference of intensity of the two lights no matter whether, 1, 2, 3, 4, or 5 glowers were used. The following table shows the result.

TABLE IX

A Table Based on the Measurements of 2700 Trails Showing the Angular Deflections at Five Different Absolute Intensities

Such constancy of quantitative results is only possible where we are dealing with purely physico­chemical phenomena or where life phenomena are unequivocally determined by purely physico­chemical condi­tions.

5. It seems difficult for some biologists, even with the validity of the Bunsen-Roscoe law proven, to imagine that the movements of the animals under the influence of light are not voluntary (or not dictated by the mysterious “trial and error” method of Jennings).231 But one wonders how it is possible on such an assump­tion to account for the fact that the angle of deflec­tion of the larva of the fly when under the influence of two lights of different intensities should be always the same for a given difference in intensity; or why the time for curvature in Eudendrium should vary inversely with the intensity of illumina­tion. It is, however, possible to complete the case for the purely physico­chemical analysis of these instincts. John Hays Hammond, Jr., has succeeded in constructing helio­tropic machines which in the dark follow a lantern very much in the manner of a positively helio­tropic animal. The eyes of this helio­tropic machine consist of two lenses in whose focus is situated the “retina” consisting of selenium wire. The two eyes are separated from each other by a projecting piece of wood which represents the nose and allows one eye to receive light while the other is shaded. The galvanic resistance of selenium is altered by light; and when one selenium wire is shaded while the other is illuminated, the electric energy (supplied by batteries inside the machine) which makes the wheels turn (these take the place of the legs of the normal animal) no longer flows symmetrically to the steering wheel, and the machine turns towards the light. In this way the machine follows a lantern in a dark room in a way similar to that of a positively helio­tropic animal. Here we have a model of the helio­tropic animal whose purely mechanistic character is beyond suspicion, and we may be sure that it is not “fondness” for light or for brightness nor will-power nor a method of “trial and error” which makes the machine follow the light.