Popular Science Monthly/Volume 43/September 1893/Recent Science III
By PRINCE KROPOTKIN.
AT one of the recent sittings of the French Academy of Sciences, Henri Moissan, whose name has lately been prominent in chemistry in connection with several important discoveries, read a communication to the effect that he had finally succeeded in obtaining in his laboratory minute crystals of diamonds. His communication was followed by a paper by Friedel, who has been working for some time past in the same direction, and has attained similar though not yet quite definite results; and, finally, Berthelot, who also was working in the same field, but followed a different track, announced that, in view of the excellent results obtained by Moissan, he abandons his own researches and congratulates his colleague upon his remarkable discovery.
The discovery is not absolutely new, and the French chemist himself mentions two of his English predecessors. Mr. Hannay obtained in 1880 some diamond-like crystals by heating in an iron tube, under high pressure, a mixture of paraffin oil with lamp-black, bone oil, and some lithium; and in the same year Mr. Sidney Marsden, by heating some silver with sugar charcoal, obtained black carbon crystals with curved edges. Besides, it was generally known that a black powder, composed of transparent microscopical crystals having the hardness of diamond, is deposited on the negative electrode when a weak galvanic current is passed through liquid chloride of carbon. But these crystals, like those of Mr. Marsden, belong to the easily obtained variety of black diamonds known as carbonados; while some of the crystals obtained by Moissan are real colorless and crystallized diamonds — the gem we all know and admire.
For industry and everyday life the infinitesimal quantities of diamond dust obtained by the French chemist may have no immediate value, and some time will probably be required before a modest-sized jewel is made in a laboratory. But the discovery has a great scientific interest, inasmuch as it is the outcome of a whole series of researches which have recently been made with the view of artificially reproducing all sorts of minerals and rocks, and which are admirably chosen for ultimately throwing new light upon the intimate structure of physical bodies. RECENT SCIENCE. 623
Moissan's method is based upon the capacity of iron of absorb- ing carbon at a high temperature and of giving it back in the shape of grains and crystals while the iron mass is cooling. When iron has been saturated with carbon at a temperature of about 2,000° Fahr., a mixture of amorphous carbon and graphite is discovered in the iron mass. At higher temperatures the fused iron dissolves more and more carbon, and the cast iron of our blasting furnaces, after having been heated to about 3,000° and slowly cooled down, contains, as known, an abundance of graphite crystals. It was thus natural to see whether a still higher tem- perature, and cooling under high pressure, might not give the still denser form of carbon — that is, the diamonds.
In order to thoroughly saturate iron with carbon at a high temperature, and to cool it under a high pressure, Moissan re- sorted to a very simple and effective means. He took a hollow cylinder of soft iron, filled it with some purified sugar charcoal, and corked the cylinder with an iron screw. Then about half a pound of soft iron was molten in a crucible in Moissan's new elec- tric furnace, which readily gives a temperature of about 3,000° C. (5,400° Fahr,), and the cylinder was plunged into the molten metal ; iron was thus thoroughly saturated with carbon. The ■crucible was then taken out of the furnace and plunged into a pail of cold water until the surface of the iron mass was cooled to a dull red temperature, whereupon it was taken out and left to cool in the air. This was the ingenious means of obtaining a high pressure. It is known that water when it becomes ice increases in volume, and that if it freezes in a strong shell the Interior pressure of the crystallizing water often bursts the shell ; but if it can not burst the shell it necessarily solidifies under an immense pressure, due to the molecular forces. The same was done by Moissan with the liquid iron, which also has the proj^erty of increasing in volume while it solidifies. An outer solid crust having been formed by a sudden immersion into cold water, the crust prevents the further expansion of the iron mass, which is thus bound to solidify under an immense pressure, like the water in the shell.
The next step was to separate the iron from the carbon crystals which it might contain. This was done by dissolving the iron in hydrochloric acid, and three different varieties of carbon crystals (which are not attacked by the acid) were received as a residue. Some graphite, some chestnut-colored, curved needles of carbon, and diamond dust could be seen ; and they were separated from each other by several complex operations indicated by Berthelot in one of his previous works. A few grains of diamond dust were finally obtained — most of them belonging to the carbonado va- riety, while a few of them proved to be real diamonds ; they were
�� � 624 T^^ POPULAR SCIENCE MONTHLY.
translucent, tliey scratclied a ruby, and they distinctly showed under the microscope the crystalline structure and cleavage of the diamond ; their density was that of the precious gem, and they were completely consumed in oxygen at a temperature of 1,890°.*
Mr. Marsden's experiment with silver was also repeated; but silver being a bad dissolvent for carbon, even at a high tempera- ture, it was boiled for some time with sugar charcoal in the fur- nace, the cooling being operated in the same way as with iron. The result was extremely interesting. No diamonds were ob- tained, but a series of carbonados of different densities (from 2'5- to 3*5 times heavier than water) were discovered, some of them in grains, some others in needles, or in conchoidal masses, the densest ones also scratching ruby and burning in oxygen at 1,800°. This is perhaps the most interesting part of Moissan's re- searches, as it confirms the long-since suspected fact that there is a whole series of carbon molecules each of which is composed of a different number of atoms, and some of which must be very- complex.
As to the quantities of diamond dust obtained in this way^ they were extremely small. Several cylinders gave no diamonds at all, and from all his experiments Moissan could not collect even a few milligrammes (a few hundredth parts of a grain) of the precious dust, although the l)lack carbonados were quite com- mon. But a sure method is now indicated, and its further devel- opment is only a matter of time and perseverance.
The scientific value of these researches is undoubtedly very great. Diamond, like graphite and simple charcoal, is pure car- bon, but all attempts at fusing carbon or dissolving it have hith- erto failed ; it could not be brought into a liquid condition out of which it afterward might crystallize. However, the investiga- tions recently made into the carburization of iron, especially by Roberts Austen, tended to prove that in steel and cast iron the carbon is not simply diffused through the iron, but enters with it into some of those combinations in definite proportions which^ like all solutions, occupy an intermediate position between real chemical compounds and purely physical mixtures, f It was reasonable, therefore, to presume that carbon is brought into a liquid condition in molten iron, and that under certain conditions it may crystallize in the shape of diamonds within an iron mass. Moissan's discovery confirms this view. On the other side, the researches of Moissan and Fried el must also throw some light
- PVoiii a suhsefiuent comiiiiiiiifiition by Moissan we learn that tlie same varieties are
found in the diarnonfl-l)eaiinf: eaith at the Cape.
f See Recent Science, in Pojiiilar Science Monthly, October, 1892.
�� � RECENT SCIENCE. 625
upon tlie great questions raised by MendeMeff as regards the prob- able presence and prevalence of iron and carbon compounds in the interior of the globe, the formation of naphtha out of these compounds, and other extremely interesting geological questions.*
The artificial reproduction of the diamond must also be viewed as a further step in a long succession of researches which have been lately pursued for artificially reproducing all sorts of min- erals, the formation of which had long remained a puzzle for mineralogists. The silicates which were formerly considered as impossible to reproduce in the laboratory have yielded within the last few years before the efforts of the chemists. Sarrasin, Haute- feuille, and especially Friedel, have reproduced different varieties of the chief constituent mineral of our crystalline rocks — feldspar — and the artificial crystals are absolutely identical with those found in Nature. Hornblende, which had long defied the efforts of the explorers, has been finally obtained in 1891 by K. Chrust- choff, after he had spent seven years in unsuccessful attempts ; f but in order to reproduce it he had to heat its constituent ele- ments for three months at a temperature of nearly 1,000°. The importance of a high temperature for further achievements was rendered still more evident in Fr^my^s successful reproduction of the ruby. The ruby is, of course, quite different from the diamond. Like the sapphire and the corundum, it is nothing but alumina — that is, a compound of two atoms of aluminium with three atoms of oxygen, colored by some impurities in red, in blue, or in brown. But for a long time alumina would not crystallize in our laboratories. Later on, Fr^my obtained a very fine dust of rubies ; but when he submitted the constituent parts of the ruby to a temperature of 2,700°, and maintained the same temperature for one hundred consecutive hours, he was rewarded by full-sized crystals of the precious stone, big enough and in sufficient num- bers to have a collar made of them. And, finally, the investiga- tion of Friedel, Le Chatelier, and especially F. Fouqud and Michel Levy, who reproduced a micaceous trachyte containing feldspar, spinel, and mica, demonstrated the necessity of resorting to a high pressure in addition to a high temperature.
To extend the range of high temperatures hitherto obtained, and to devise a means of measuring them, was thus the first con- dition for further progress in the reproduction of minerals and gems. But the measurement of high temperatures is a very diffi- cult problem which has much occupied of late several prominent physicists and chemists. A thermo-electric thermometer, made of two very resistant metals (platinum and an alloy of platinum
��* See, in Mendeleeif' s Principles of Chemistry, the footnotes to the chapters on carbon and iron. \ Comptes Rendus, 1891, tome cxii.
VOL. XLIII. — i5
�� � 626 THE POPULAR SCIENCE MONTHLY.
with rhodium), and graduated with the aid of the air ther- mometer, finally came into general use, and it proved to be quite reliable — but only up to 3,000° Falir.,* which temperature was soon surpassed. Then, Le Chatelier devised a pyrometer based on the variations of intensity of light of fused metals at different temperatures, and this instrument again proved to be sufficiently accurate up to 3,600° ; but this last temperature, too, is now sur- passed by Moissan, by means of his new electric furnace, which is a real model of efficiency and simplicity, f It consists of two superposed bricks, made of quicklime, or of an especially pure calcinated magnesia. A groove with a small cavity in its middle (large enough to receive a small crucible) is made on the upper face of the lower brick in the sense of its length, and two carbon electrodes are introduced from both sides into the groove. As soon as they are connected with a dynamo machine the electric arc appears between their extremities, and an immensely high temperature is produced in the cavity. Thus, a small Edison machine, worked by a gas engine of eight horse power, gave a temperature estimated at" about 4,500° Fahr., and with a fifty- horse-power engine the enormous temperature of about 5,400° (3,000° C.) was reached.
The effects of this little furnace are simply wonderful. At about 4,500° lime, strontia, and magnesia are crystallized in a few minutes. At 5,400° the very substance of the bricks is fused and flows like water. Oxides of various metals which were considered as quite irreducible are deprived of their oxygen in no time; nickel, cobalt, manganese, and chrome oxides can be reduced at a lecture experiment, and a piece of 120 grammes of pure uranium is obtained at once from the uranium oxide. At about 4,050° pure alumina is fused and little rubies are formed ; true, they are less beautiful than those of Frdmy, but the whole experiment lasts less than a quarter of an hour. At a higher temperature alumina is even volatilized, and nothing is left of it in the crucible. In short, the results are as interesting and as promising as those which Pictet and Dewar have witnessed when they went to the other end of the thermonietric scale and produced the extremely low temperatures of about 200° C. below the freezing point.
And, finally, Moissan's discovery establishes a new link be- tween the processes which we obtain in our laboratories and those which are going on in the celestial spaces, in the formation of meteorites. It was known long since that these masses of silicates and nickeled iron which travel in the interplanetary
- C. Bams in Philosophical Magazine, fifth series, xxxiv, 376 ; L. Holbom and W. Wien
n Wiedemann's Annalen, xlvii, lOY.
f Comptes Rendus, December 12, 1892, tome cxv.
�� � RECENT SCIENCE. 627
spaces and, entering occasionally into the sphere of attraction of the earth, fall upon its surface, sometimes contain charcoal or a special variety of graphite ; but later on, in 1887, the St. Peters- burg Professors Latchinoff and Erof^eff went a step further and proved that the charcoal is occasionally transformed into diamonds ; thus they extracted some diamond dust from the meteorite fallen during the previous year at Novo Urei, in the province of Penza. Some doubts were, however, entertained as regards their discovery, but the fact has been fully confirmed since by Friedel and Le Bel, who found in a meteorite from Canon Diablo minute diamonds and carbonados exactly similar to those of Moissan.*
It is thus evident that the artificial reproduction of the diamond is not one of those accidental discoveries which may be made without leaving an impression upon science for many years to come. It is only one of the many advances made in a certain direction, and is the outcome of the whole drift of modern research which endeavors immensely to widen the means at our disposal for effecting physical and chemical transformations of matter. It is one step more into a new domain where chemistry, metal- lurgy, and mineralogy join hands together for revealing by joint efforts the secrets of the constructive forces of matter.
The study of the direct action of environment upon organisms, and of the mechanism of its action, becomes a favorite study among biologists — the " transf ormists " being no more a few ex- ceptions in science, but already constituting a school which has several brilliant representatives in America, France, and Germany, as well as in this country. It is evident that almost none of the biologists engaged in this kind of research maintains any doubts as to the importance of natural selection as a factor of evolution. To use the words of one of the leading American transf ormists, f "the law of natural selection is well established, and no more under discussion." For many adaptations it offers the best and the only possible explanation. But biology would have been brought to a standstill if the idea had prevailed that, after a more or less plausible explanation of some adaptation has been given under the hypothesis of natural selection, nothing more is left to be done to explain this same adaptation. For many animals whose manners of life we hardly know at all — the study of animal life having been deplorably neglected for the last fifty years — the ex- planation would often be little better than a mere hypothesis ; but
- Comptes Readus, December 12, 1892, tome cxv, p. 1039; also February 13, 1893.
f H. F. Osborn, whose admirable essays, mentioned in a previous review, are now pub- lished in book form.
�� � 628 THE POPULAR SCIENCE MONTHLY.
even in the best cases the very origin of each variation would still remain to be found. Darwin fully understood this necessity ; and the physiological and mechanical origin of variations is what so many biologists are now working at. Several such investigations are already well known to English readers through the works of Cope, Semper, Lloyd Morgan, J. T. Cunningham, and P. Geddes, Many others ought to be analyzed and discussed ; but for the time being I can only mention a few recent works relative to the origin of animal colors.
Wherever we go we see animals colored in accordance with their surroundings. White and light gray colors predominate in the arctic regions ; tawny and yellow colors in the deserts ; gor- geous colors in tropical lands. The striped tiger in the jungle is hardlj'- recognizable among the shadows of the tall grasses. Insects resemble the flowers which they usually visit ; caterpillars have the colors and often the forms of the twigs and the leaves they feed upon. Dusty-colored nocturnal insects; moths which take autumnal tints if they begin life in autumn ; dark squirrels in the dark larch forests, and red squirrels in the Scotch-fir groves; animals changing their color with the season — all these are famil- iar instances. But are they all due to natural selection alone ? Does not environment take some part in itself producing these colors ?
In a very suggestive work*, Alfred Tylor has shown in how far the different markings and the diversified coloration of animals follow the chief lines of structure ; and A. R. Wallace has readily admitted that, while the fundamental or ground colors of animals are due to natural selection, the markings are probably due to internal physiological causes.f Coloration responds to function ; and there is a law in the distribution of colors and the develop- ment of the markings, while there ought to be none under the hypothesis of selected accidental variations. Wallace goes even a step further, and shows that those birds possess the most brilliant colors which have developed frills, chests, and elongated tails, or immense tail-coverts, or immensely expanded wing feathers, all appearing near to where the activities of the most powerful muscle of the body would be at a maximum. He considers " a surplus of vital energy," increased at certain periods, as a vera causa for the origin of ornamental appendages of birds and other animals. And it is difficult to examine these and like facts with- out coming to the same conclusion.
But if partial vigorous coloration is so much dependent upon vital energy, is it not possible to suppose that the decoloration of animals with the approach of the winter is in some way connected
- CJoloration in Animals and Plants. London, 1886. \ Darwinism, p. 288 et seq.
�� � RECENT SCIENCE. 629
witli a decrease of vital energy, especially if we take into account the permanent wMte colors of domesticated animals in arctic regions (such, as the Yakutsk horse), which can not be dependent upon natural selection ? Some recent observations give a certain support to this supposition. Thus we now learn that rabbits which have been taken to the Pic du Midi Observatory (9,500 feet above the sea level) have given in seven years a race somewhat different from their congeners in the surrounding plains. They are a little smaller, have less developed ears, and their fur coats are of a lighter color and very thick. Moreover, the very consist- ence of their blood has undergone a notable change. It contains more iron, and possesses a greater power of absorption for oxygen.* An anatomical change is thus produced by the environment ; and no naturalist will doubt that, if the race continues to multiply for a great number of years in the same conditions, it will maintain its present characters or develop new ones on the same lines, the more rapidly so if natural selection eliminates the less adapted individuals.
A few more additions in the same direction may be found in a valuable work recently published by F. E. Beddard.f Thus, he mentions the researches of Dr. Eisig, X who has endeavored to explain the ground colors of some animals as dependent upon their food, and has shown, for instance, that the yellow color of an annelid which is living on a yellow marine sponge (a color which might be explained as protective for the parasite) depends upon the yellow pigment of the sponge absorbed by the annelid. The prevalence of crimson colors among some fishes in a certain part of the New England coast, which is covered with scarlet and crimson seaweeds, is explained by J. Browne Goode by the red pigment derived by the crustaceans from the algae with which their stomachs are full, the crustaceans being devoured by the fishes. And the experiments of Mr. Guyson relative to the effects of different food plants ujDon a number of species of moths, as well as those of Mr. J. Tawell upon important modifications produced by food in the larvae of the large tortoise-shell butterfly, both mentioned in the same work, are attempts in a most im- portant but very young branch of experimental morphology.
Another series of researches is now being made with the view of more deeply penetrating into the physiological causes of animal coloration. Thus, it is a fact well known to fishermen, and now
��* Comptes Rendus, January 2, 1891, tome csii.
■j- F. E. Beddard, Animal Coloration ; an Account of the Principal Facts and Theories relating to the Colors and Markings of Animals, London, 1S92.
X Fauna und Flora des Golfes von Neapel : die Capitelliden, quoted by Mr. Beddard, loc. cit., p. 101.
�� � 630 THE POPULAR SCIENCE MONTHLY.
confirmed by direct experiment, namely, by Westboff, that several fresb-water and marine fisbes cbange tbeir color from wbite to dark as soon as tbey bave been transferred from a medium witb a ligbt-colored bottom to anotber medium the bottom of wbicb is dark. Fisbermen, we are told by Mr. Poulton, even keep their bait in wbite-colored vessels in order to make it assume a ligbter color. The common frog also can cbange its color to some extent in barmony witb its surroundings, wbile the green tree-frog of southern Europe was long since known for tbis capacity. It is bright green among green leaves, and dark green when seated on the earth or among brown leaves.* Like changes are also known in the chameleon and in some South American lizards. The causes of these changes have already been investigated by Pouchet in 18-48 and Briicke in 1852, but now we have a more elaborate research by Biedermann f upon the same subject. He has dis- covered three different layers of cells which contribute to give the frog its varying colors. There is first, deeply seated in the skin, a layer of pigment-cells which contain black pigment both in their interior and in their ramified processes, spreading within the skin. These cells are covered by a second layer of " interfer- ence-cells " containing bright yellow granules as well as granules of a pigment which sometimes appear blue or purple, and some- times gray — the whole being covered with a transparent outer skin. The normal green color of the frog is produced by a com- bination of blue and yellow interference-cells appearing on a black background ; but if the black pigment of the deepest layer is protruded into its ramifications, the color of the animal becomes darker ; and if it retires deeper, the yellow granules of the middle layer become more apparent, and the frog assumes its lemon- yellow color. Finally, when the yellow pigment gathers into round drops between the bluish interference-cells — not above them — the skin acquires a whitish-gray tint. The same arrange- ments exist in other reptiles and amphibia.
Now, how is it that the cells cbange tbeir position in various lights ? Is it some reflex action in the nervous system, as it appears in fishes, which cease to change their color when tbey become blind ? Or have we to deal with some direct action of light ? Facts are in favor of the second explanation. The slight- est change of temperature affects the mutual disposition of the pigment-cells, and consequently the color of the frog ; it is enough to keep the animal in the hand to provoke a contraction of its black cells. The amount of blood-supply also has a definite effect ;
- E. B. Poulton, Colors of Animals, London, 1890, p. 82 et seq.
f W. Biedermann, Ueber den Farbenwechsel der Frosche, in Pfliigei's Arcbiv f iir Physi- ologic, 1892, Bd. li, p. 455.
�� � RECENT SCIENCE. 631
as soon as a certain part of the skin receives no more blood, the color-cells receive less oxygen, the black cells contract, and the animal assumes a lighter color. But the effects of light are even more interesting. Pouchet had shown that those fishes which usually adapt their color to their dark or light surroundings cease to do so when they have lost sight ; they remain dark even in light surroundings.* The indirect effects of light through the intermediary of the visual organs are thus certain. But Steinachf has proved that light acts in a direct way as well — perhaps, we may add, in the same way as it acts upon the chlorophyll grains of the leaves. He glued strips of black paper to the skin of frogs which were kept in the dark, and when these animals were exposed to light, only the open parts of their skin returned to a lighter color, while the covered parts remained dark. To avoid all doubts, the experiments were repeated on skin separated from the body, and photograms of letters and flowers, cut out of black paper and glued to the skin, were reproduced upon it. Besides, iDlind tree-frogs do not darken as the fishes do, and Biedermann has proved that the chief agency of their changes of color is not in the sensations derived from the eye, but in those derived from the skin. Frogs, whether blind or not, become dark green, or black, if they are kept in a dark vessel in a sparingly lighted room. But when a larger branch with green leaves is introduced into the vessel, they all recover their bright-green color, whether blind or not. In some way unknown, the reflected green light acts either upon the nerves of the skin, or, what seems more prob- able, if Steinach's experiments are taken into account, directly upon the pigment-cells. Moreover, the sensations derived from the toes have also an influence upon the changes of color. When the bottom of the vessel is covered with felt, or with a thin wire net, the frogs also become black, recovering their green color when a green branch is introduced in the vessel.
We have here temporary changes of color produced by the surroundings ; but various gradations may be traced between the temporary and the permanent changes. Thus Lode provoked local contractions of the pigment-cells in fishes by electrical irritations applied locally. And Franz Werner's researches upon the color- ing of snakes, recently embodied in a separate work, J show that the temporary and irregular spots which appear in fishes and frogs under the influence of artificial irritations are of the same
��* Direct observations have been made also by Alois Lode (Sitzungsbericbte of the Vienna Academy, 1890, vol. xcis, 3te Abtheilung).
+ Ueber Farbenwechsel bei niederen Wirbelthieren, bedingt durch directe Wirkung des Lichtes auf die Pigmentzellen, Centralblatt fiir Physiologic, 1891, Bd. v, p. 326.
\ Franz Werner, Ueber die Zeichnungeu der Schlangen, Wieu, 1890.
�� � 632 THE POPULAR SCIENCE MONTHLY.
character, and have the same origin, as the also temporary and irregular spots which appear in other fishes, as well as in several tritons and many Gekonides, without the interference of man. Some of the provoked changes of color do not entirely vanish after the irritation is over, and they belong to the same category as the spots which appear in many animals in youth, and disap- pear with growing age. Moreover, it is maintained that a series of slow gradations may be established between the irregular spots, the spots arranged in rays, and finally the stripes, such as we see them in higher mammals like the zebra or the tiger ; and if these generalizations prove to be correct, we shall thus have an un- broken series, from the temporary spots provoked by light or electricity to the permanent markings of animals.*
And, finally, attempts are being made to explain some of the wonderful so-called adaptive colors of insects as a direct product of environment. Some time ago (in 1867) T. W. Wood published experiments upon the larvae and pupse of both the small and the large cabbage butterfly. He kept the larvae during their meta- morphoses in boxes lined with paper of different colors, and he found that the colors assumed by the pupae more or less corre- sponded to their surroundings. Later on E. B. Poulton made a wider series of analogous experiments, and he saw that the change of color is accomplished during the first hours when the larva spins its web ; he came to the conclusion that it depends upon a certain physiological action which is transmitted to the nervous system, not only through the visual organs, but through the whole surface of the skin. These facts have now been fully con- firmed again by W. Petersen, f but his explanation is of a more mechanical character. He maintains that the color of the pupa depends upon the pigment contained in both its cuticle and hypodermis. The pigment of the latter is green in the larva, and sometimes it remains green during the pupal stage ; but it may be visible or not, according to the amount of dark pigment which is formed in the cuticle, and the amount of this dark pigment entirely depends upon the color of the light. Yellow and orange light prevents the formation of the dark pigment, and in such cases the cuticle, which remains transparent, shows the green pigment of the hypodermis. But the less bright parts of the spectrum have not the same power, and if we trace a curve rep- resenting the powers of the various parts of the spectrum for preventing the formation of a dark pigment, the curve has its
��* See the polemics engaged upon this subject in Biologisches Centralblatt, December 15, 1890, and July 15, 1891 ; as also the Zoologische Jahrbiicher, 1891.
•)• Zur Frage der Chromophotographie bei Schmetterlingen, in Sitzungsberichte der Dor- pater Naturforscher-Gesellschaft, 1890, vol. x, p. 232.
�� � RECENT SCIENCE. 633
culminating point in the yellow, and descends toward both ends of the spectrum ; it exactly corresponds with the curve of assimi- lation of carbon by plants under variously colored light. It is also remarkable that the green color of the pupa is only obtained by yellow light, or by such green as contains yellow ; such is, as known, the average color of leaves. We thus have a case where en- vironment itself makes the color which approximately matches it. The meaning of these researches is self-evident. No naturalist will probably attempt to explain the animal colors and markings without the aid of natural selection. But it becomes less and less probable to admit that the animal colors are a result of a selection of accidental variations only. The food of the organism, and especially the amount of salt in it, the dryness or moisture of the air, the amount of sunshine, and so on, undoubtedly exercise a direct effect on the color of the skin, on the fur, and on the very intimate anatomical structure of the animal. As to the relative parts which must be attributed in the origin of each separate variation to natural selection on the one side, and to the direct action of environment on the other side, it would simply be un- scientific to trench uj)on such questions in a broadcast way, so long as we are only making our first steps in discriminating the action of the latter agency. The first steps already indicate how complicated such questions are, especially in those cases where natural selection must act in an indirect way — not as a mere selection of already modeled forms, but as a selection of forms best capable to respond to the requirements of new conditions — in which case the intimate organization of the living being comes in the first place. All we may say at the present moment is that the direct modifying action of environment is very great, and that no theory can claim to be scientific unless it takes it into consideration to its full amount. — Xineteenth Century.
��Mr. W. Roe, of the Cape Colony, has pointed out a disadvantage connected with irrigation. Most water nsed for the purpose contains variable (quantities of soluble salts, some of which are not taken up largely by plants. Every applica- tion of water, therefore, adds to the saline ingredients of the soil — a very differ- ent effect from that of excess of rain water, which, so far as there is open subsoil for it to drain away, would be likely to take out rather than add to the soluble salines in the soil. The mischief of the accumulation of salts in the soil is aggra- vated in a dry-air land where evaporation is great. The air, acting like a sponge on a surface, takes up the water, leaving the accumulated salts in the surface soil. But this surface soil is as the sponge to the layer beneath. Constantly after each water-leading, the water is drawn to the surface and evaporated, leaving the ac- cumulated salts in the surface soil. The harm done by this accumulated salt will depend on the nature and quantity of the salines in the water used, as also upon the quantity of water supplied. VOL. XLin. — 46
- Comptes Rendus de l'Académie des Sciences, February 6, 1893, tome cxvi, p. 218.
- Proceedings of the Royal Society, xxx, 188; quoted by Moissan.
- Proceedings of the Royal Society of Edinburgh, 1880, ii, 20 (Moissan's quotation).