Hiram?" said the old lady. "Why, now I can tell when it is going to rain!" "Hiram," said she, "I am truly ashamed of you. What do you suppose the good Lord gave you your rheumatism for?" Aerographers are in a way like the old lady. We would like to ask the Smithsonian investigators "What do they think the good Lord gave us an atmosphere for?" This great thermal engine, so well described by Sir Napier Shaw in his last paper on "The physical structure of the atmosphere." Are we no longer to study from a dynamical point of view the transfer and transformation of energy? The work of years on the convergence of air streams, the development of areas of turbulence and discontinuity-is all this to be relegated to the rear and an undetermined variation in the value of the solar constant of radiation outside the atmosphere to be given first place in forecasting weather?

Dr. Abbot tactfully omitted reference to the serious side of the review, which was to the effect that the results given in the publications seemed to the reviewer to have no substantial basis as factors of value in forecasting. The results negative the claims made.

Dr. Abbot says that perhaps I failed "to see the forest because it was obscured by the trees." Oh! well! I was not looking for forests or even trees, only searching with wistful eye for some sign of vegetation in a dry and thirsty land.

BLUE HILL OBSERVATORY

ALEXANDER MCADIE

AN APPEAL FOR SIMPLIFIED LITERATURE

CITATIONS

THE average editor and the average publisher of scientific literature is ultra-conservative with respect to the form of reference citations, whether these be in terminal bibliographies, footnotes or text references to places of publication of technical names. This same statement may also apply equally well to the average author of scientific papers; although usually the author has no choice, being obliged to follow established usages no matter how antiquated and cumbersome these may be, in order to conform to editorial mandates. Forms adopted many years ago are currently followed and in many cases little or no attention is given to atility or to simplification. In some scientific periodicals, the use of the cumbersome Roman numerals for indicating volume numbers has been abandoned, but in a very high percentage of modern periodicals, including a number of recently established ones, this ancient form is still used. It may be doubted if editors, publishers or authors give much attention to these seemingly small details, being

more apt to follow the line of least resistance and past or current usage.

The conciseness, clearness and utility of indicating the volume number by black-faced figures 38 as compared with the Roman characters XXXVIII is manifest. With the use of the black-faced figures it is not necessary to stop to translate, as is frequently the case when the cumbersome Roman system is used. We are all more or less familiar with the lower numbers in the Roman series from long usage, but few can rapidly translate the more complicated higher figures. In this age of rapid publication shall we go to the extreme, when called upon to cite such a publication as Bulletin 1348 of the U. S. Department of Agriculture by translating the simple figures into the cumbersome MCCCXLVIII? In a larger number of modern serial publications this absurd procedure becomes necessary because of usage and established editorial custom and is pedantic in the

extreme.

There is little uniformity in scientific literature in reference to the form of citation. It not infrequently happens that an author in preparing a paper for submission to a certain journal follows the form approved for that serial, but if he changes his mind and later desires to submit the paper to some other serial, he frequently has to rewrite considerable parts of the manuscript in order to bring it into conformity with the style followed in the second one. This is especially true if the paper happens to be a taxonomic one with numerous literature citations or one with an extensive bibliography.

In several modern standard review publications the conciseness and utility of the simplified form of citation has been amply demonstrated, but editorial usage in review publications has had little or no influence on the forms used in established technical periodicals, while the editorial staffs of newly established serials frequently give no consideration to the matter. We are all familiar with review literature, but most of us are impervious to the manifest advantages of the simplified citation forms adopted by the majority of them. The simplified form adopted by several standard review publications is given below:

Chemical Abstracts: 24, 57-70 (1925).
Science Abstracts: −24. pp. 57–70, Feb., 1925).
Botanisches Centralblatt: 1925, 24, 57-70.
Botanical Abstracts: 24: 57-70. 1925.

In order to indicate the wide range of variation in reference citation, even where the Roman system of indicating volume numbers is not used, the following data have been compiled from the same reference in four different periodicals:

(1) Experiment Station Record: (Ann. Appl. Biol., 24 (1923), No. 2, pp. 151-193, pls. 3, figs. 31).

(2) Journal of Agricultural Research: In Ann. Appl. Biol. v. 24, 1923, p. 151–193, pl. 1-3, fig. 1-31.

(3) Philippine Journal of Science: Ann. Appl. Biol. 24 (1923) 151-193, p. 1-3, fig. 1-31.

(4) Botanical Abstracts: Ann. Appl. Biol. 24: 151-193. 3 pl. 31 fig. 1923 [or pl. 1-3, fig. 1-31. 1923].

Number of Characters

56

52

45

40-46

In No. 1, with 56 characters, no data except the irrelevant "No. 2" are given that are not included in the shortest form utilizing but 40 characters. Two pairs of parentheses, "No. 2," "pp." and several commas are redundant. There is no differentiation in type; that is, nothing to catch the eye.

In No. 2, with 52 characters, the following are redundant: "In," "p," "v" and several commas. There is no differentiation in type.

In No. 3, with 45 characters, the parenthesis is redundant, owing to the place of the date of publication. There is, however, proper differentiation of type as to volume, plates and figures.

In No. 4, with 40 characters in its simplest form, there are no redundant letters, figures or punctuation marks, the latter being reduced to a single colon and several periods. The volume number, page, plate and figure references are properly differentiated. Nothing essential is left out. It is the easiest to read and to proof read; the easiest to write, whether long hand or on the typewriter; and what is still more important presents a minimum chance of error. These points are perhaps matters of slight importance in short papers having only a few references but become of very great importance in those works having hundreds and even thousands of ref

erences.

There is little force in the argument that type of different styles such as black face, Roman and italics should be avoided in the same line. Most modern composition work is done on the linotype or monotype machine and with these machines the use of different fonts is practically as simple from the standpoint of the compositor as it is for a copyist to operate the shift key on a typewriter for upper case characters.

Merely because an established form of citation has been followed for many years is no reason why a change should not be made, especially if the change still makes the reference entirely clear and eliminates useless characters. Utility, simplicity, clarity and brevity should be the criteria, not past or current

custom. The general adoption of the concise form utilized in Botanical Abstracts by publishers, by responsible editors of technical literature and by authors of technical papers is greatly to be desired. This appeal for the simplified form of citation is primarily directed to publishers, editors and members of editorial staffs. Unless the initiative be taken by these, the individual author is powerless in the matter. Scientists are frequently accused of not being practical, but here is an opportunity of demonstrating on a small scale a distinctly time-saving device that would in the long run make our published data simpler, clearer and more attractive. I venture the prediction that no author who has once prepared a paper in which the simplified form of citation herein discussed is used will voluntarily revert to the more ancient complicated forms that still prevail in the majority of our technical publications, whether these be in serial form or individual volumes. E. D. MERRILL

UNIVERSITY OF CALIFORNIA

THE EFFECT OF MINERAL SUPPLEMENTS ON REPRODUCTION OF THE ALBINO RAT

FAILURE of growth in the second generation of rats on a ration of corn and peanut meal supplemented with acid phosphate was reported last year.1 The basal ration was later fortified with dried meat and cod liver oil. When the improved ration, consisting of white corn 30, wheat 30, peanut meal 25, dried meat 12, and cod liver oil 3, was supplemented with 1.0 per cent. of NaCl and 1.5 per cent. of acid phosphate, reproduction was not more successful than on the acid phosphate ration in the former experiments. The females on this ration farrowed normal numbers of young; but they apparently furnished very little milk and the pups soon became thin and scrawny. Many of the pups were eaten by the dams. None of the litters reached the age of weaning.

The addition of sufficient NaHCO, (0.5 per cent. of the total ration) to neutralize the excess of H2SO carried by the phosphate resulted in marked improvement. We now have young rats, representing the third generation on this ration, that are being raised successfully.

Other investigators have shown that the pig, goat, cow, rabbit and monkey are able to combine ingested mineral acids with ammonia and excrete the ammonium salts through the urine. Some of their experiments have shown that protein storage on a normal level of protein intake or growth through a considerable period of time was not interfered with. It has not been shown, however, that normal repro1 Salmon, SCIENCE, 60, 457, 1924.

Our observations show that a small amount of H2SO, may limit the growth of the rat in the second

generation even when little or no effect on the growth of the first generation was apparent.

A study of the effect of various acids and acidforming substances on the reproduction of rats and swine is now being made in this laboratory. W. D. SALMON

LABORATORY OF ANIMAL NUTRITION, ALABAMA POLYTECHNIC INSTITUTE

THE HYDROID OF CRASPEDACUSTA
RYDERI IN KENTUCKY

SINCE the finding of the hydroid of Craspedacusta ryderi1 (1920) in Boss Lake, Elkhart, Indiana, I have been anxious to find it elsewhere, especially so since all the medusae at Elkhart were females and all medusae found elsewhere were males. I have been interested in the sex question involved and wish to transplant hydroids from some other source to Boss Lake.

July 30 I made a trip to Benson Creek, where Garman2 found the medusae in 1916, 1917 and in 1924, and was rewarded by finding the hydroids on the rocks in the shallow water just above the place

where the medusae were most abundant. As the water was muddy and most of the rocks covered with slime only a few hydroids were found. Some of these I took to my laboratory, where they now are and where I hope to be able to rear them. When the water clears I shall try to find more of them.

timely interest. The present volume is practically a republication of the former work, with the addition of three new chapters on special subjects and an appendix of personal letters. It makes a handsome large octavo volume of 773 pages, with 36 page plates and 130 text illustrations. The illustrations are slightly less in number than in the former volume. Twenty of the plates and 105 of the text figures are portraits. This is the first work published on the Philip Hamilton McMillan Memorial Publication Fund.

The history covers the ten decades from 1785 to 1885, grouped in eight eras. The first two eras are named after the two most active workers in the early days, William Maclure, the "father of American geology," and Amos Eaton, the earliest teacher of the science. The Maclurean Era includes the years 1785–1819, and the Eatonian 1820-1830. Five chapters cover five decades of State Geological Surveys (1830–1888), with chapter eight devoted to the National Surveys. The special problems discussed in the previous work were, "The fossil footprints of the Connecticut Valley," "The Eozöon question," "The Laramie question" and "The Taconic question." The three new special esis"; "The development of micro-petrology," and topics are, "The development of the glacial hypoth

"How old is it?"

The biographical sketches of the geologists are not the least interesting part of the history, and the author has described the workers and their work with discrimination and fairness. The appended personal correspondence is a welcome addition. It also em

INDIANA UNIVERSITY

FERNANDUS PAYNE

SCIENTIFIC BOOKS

The First One Hundred Years of American Geology. By GEORGE P. MERRILL, head curator of geology, United States National Museum. XXI + 773 pages, 62 x 10 inches. Yale University Press. 1924. IN the report of the U. S. National Museum for 1904 Dr. Merrill published "Contributions to the his1 Payne, Journ. Morph., Vol. 38.

2 Garman, SCIENCE, Vols. 44, 56 and 60.

3 Three weeks later hydroids were found in abundance and transplanted to Elkhart.

the pioneer work. Geology is not an exact science, but relies on observation, diagnosis, comparison and interpretation. The personal element in the early years is shown by the diverse and even contradictory views on phenomena and features which to-day are lucid. A touch of humor is suggested by the placing of the portraits of Cope and Marsh side by side.

romance.

The student of geology finds the history a real The young worker especially needs it for breadth of information and as suggestion of caution in his work and modesty in opinion. Dr. Merrill has done a good service not only to the geologic profession but to general science, and to the history of the evolution of real knowledge. It was a rare and fortunate combination for the author to have access to the literature and records, the time, patience and industry for collecting the vast mass of fact and the knowledge and discrimination necessary for its effective presentation.

The only suggested criticism of the work is its

ABOUT a year ago1 we gave a preliminary account of the effect of ionizing pure ethane by means of radon (radium emanation) mixed with it. This work has since been extended, and a further preliminary report will now be made. The gases hitherto studied are methane, ethane, propane, butane, ethylene, acetylene, cyanogen, hydrogen cyanide and ammonia; other reactions studied are: oxidation of all the foregoing except ethylene and hydrogen cyanide; the hydrogenation of acetylene, ethlene and cyanogen, and the polymerization (also in the presence of nitrogen) of acetylene, of cyanogen and of hydrogen cyanide.

In methane, propane and butane the same kind of behavior was encountered as for pure ethane. The idea has been abandoned (also for ethane) that free carbon results from these reactions. In fact, it is more exact to use the term condensation (with elimination of hydrogen) instead of decomposition. The yellow or brownish color of the liquid or solid condensates we now attribute to some unsaturation, not to free carbon.

In the case of methane the pressure change at room temperature is not a safe criterion of reaction, for the pressure remains almost constant, even when analysis shows that the greater part of the gaseous product is free hydrogen which is accounted for by reaction without volume change, such as:

The condensation of a liquid from methane is much delayed, perhaps, because it must build up in successive steps from the lowest member, methane. In propane and in butane liquid appears in droplets much sooner even than in ethane, and later may pass partly into the solid state, if a sufficient quantity of radiation be absorbed by the liquid. The phenomena in propane contained in a small volume where the density of radiation at the wall was high were particularly striking; minute gelatin-like masses with 1 SCIENCE, 60, 364, October, 1924.

sharply defined contours built up to an appreciable depth on the wall and later curled away from it; while in a larger volume with a lower radiation density only liquid appeared. In radiating methane the total pressure at 25° C. remained practically constant during the entire reaction.

In the earlier paper it was suggested that the liquid phase formed from ethane might be octane. We now know that the ratio of hydrogen to carbon in the liquid condensates is about 1.8 to 1, which would indicate some degree of unsaturation or ring compounds, or both. Analysis further shows that methane as well as hydrogen is a product of the condensation of ethane, of propane and of butane; the ratio of hydrogen to methane liberated remained constant throughout the reactions and had the following values: for ethane about 6 hydrogen to 1 methane, for propane about 4 to 1, for butane about 6 to 1. Small proportions of other saturated hydrocarbons are also found, e.g., propane and butane in the ratio of 1 to 2 from ethane; and ethane and butane in the ratio of 4 to 5 from propane. Evi

dently, if either one hydrogen or one methane molecule can be eliminated from a di-polymer of ethane (as assumed in our former paper) the residues would be butane and propane, respectively, and so forth for other similar possibilities. If two hydrogen molecules be eliminated from the di-polymer of a saturate, unsaturation would result, as illustrated by the ratio found, 1 carbon to 1.8 hydrogen (approximately the same in the liquids from ethane, propane and butane). Probably the two reactions involving the elimination of one or of two hydrogen molecules proceed simultaneously. Evidently, by this means, starting with any saturated hydro-carbon we can get both lower and higher members until soon all the members up to quite high polymers will be represented in the mixture, thus suggesting the mixtures occurring in petroleum.

In these ionic condensations of saturated hydrocarbons the average number of molecules of hydrocarbon condensing per positive hydrocarbon ion, M/N, varies from 1.4 to 2. For the unsaturated hydrocarbons M/N is higher, for ethylene (double bond) about 5, for triple bond compounds yet highercyanogen, 7; hydrogen cyanide, 10; acetylene, 20,2 the highest yet found. This shows that the ionic 2 Professor W. Mund, of the University of Louvain, kindly communicated this value for acetylene (which we have confirmed) in advance of a forthcoming publication (Bull. Soc. Chem. Belg., 34, 241, May, 1925), the appearance of which has been delayed by a printers' strike in Belgium. We have also confirmed in general the results reported by Mund and Koch (ibid., 34, 119, February, 1925) on the radiation of ethane, ethylene and acetylene.

clusters are larger for the double and triple bond compounds than for the saturated ones.

Cyanogen yielded a brownish to black solid resembling paracyanogen; a little nitrogen (about 5 per cent.) was liberated during the polymerization. Hydrogen cyanide gave a reddish to brown solid, which by analogy may be called "parahydrocyanic acid," while liberating about 5 per cent of permanent gas (which a preliminary analysis showed to be nitrogen). Ethylene yielded a liquid and quite a large volume of free hydrogen, as reported by Mund and Koch (loc. cit.) with possibly some methane. Acetylene yielded with small evolution of hydrogen (about 2 per cent. of the initial volume of acetylene) a yellow (white in the earliest stages) solid powder similar to "cuprene" -also in agreement with Mund and Koch.

The oxidation of methane and of ethane proceeds completely to carbon dioxide and water at a rate satisfying the general kinetic equation previously developed for hydrogen oxidation, which is also applicable to carbon monoxide oxidation, and to many other reactions. The principle of the exclusivity of

oxidation exhibited in the oxidation of carbon monoxide by oxygen (J. A. C. S., Nov., 1925) applies strikingly to both methane and ethane; e.g. methane in the presence of oxygen does not liberate hydrogen as it does alone, and ethane neither liberates hydrogen, nor does liquid phase appear, except water which is easily distinguished (by appearance and freezing point) from the oily liquid droplets due to its condensation to form liquid hydrocarbons. The value M(CH ̧+02) is 4.4, but in for methane oxidation of N(CH1+02) the presence of 1 molar per cent. of Se(C2H), attained a value of 5.7, only slightly below the value 6, obtained for carbon monoxide. The value for ethane M(C2H2+02) N(C2H2+02) oretical 9.0. The oxidation of propane is not complete to carbon dioxide and water in the earlier stages of the reaction; less oxygen disappears than required; oily droplets appear indicating partial oxidation products. This becomes yet more pronounced for butane oxidation; much less oxygen disappears per butane molecule than would correspond to complete oxidation; water and oily droplets appear and the carbon dioxide formation represents but a small part of the carbon leaving the gas phase.

6

=

7, is also somewhat lower than the the

The other cases of oxidation which have been studied include that of cyanogen and of acetylene. In both cases exclusively of oxidation is most striking. Acet8 Sabatier and Sanderens, Comp. rend., 130, 250 (1900); H. Alexander, Ber. deut. Chem. Ges., 32, 2381 (1899).

2

2

ylene gives a clear colorless liquid and none of the yellow powder (as it does alone); it combines with oxygen in a 1 to 1 ratio; little or no carbon dioxide is formed (except by secondary action upon the product on the wall). The oxidation product is one of direct addition (a polymer of CHO) which we have M(C2H2+02) not identified. The ratio is not less N(C2H2+02) than 16 and may be as high as 20, the value for acetylene alone. The oxidation of cyanogen gives exclusively a white powder (instead of the black polymer obtained from pure cyanogen) which is an addition product of the formula (CNO),, not described in the literature. Its properties have not yet been studied. M(C2N2+O2) The ratio is approximately 7, the same as for the polymerization of cyanogen, but some gaseous nitrogen and carbon dioxide are formed as the result of a split reaction. Recently we have effected the hydrogenation of acetylene and of cyanogen. The reactions are not exclusive, thus showing a marked difference from the case of oxidation, by which is meant that besides that part of the reaction which resulted in the addition of hydrogen to the unsaturated compounds, polymerization of the latter also occurred in a parallel reaction. This result had been anticipated from the general theory of the exclusivity of oxidation.

2

N(C2N2+O2)

2

The addition of nitrogen to acetylene was attempted in a 1 to 1 mixture. No combination with nitrogen was effected. The only reaction was that of the polymerization of acetylene to give the same yellow product obtained from acetylene alone. However, it was found that although nitrogen does not react itself, its presence much enhances the rate of acetylene polymerization; the relative rate continuing to increase owing to the increasing ratio of N2/C2H2 as the reaction proceeds. This is a new kind of catalysis which may be called ionic catalysis. It is the more striking since it was not observed in the previous cases studied. The presence of a gas which itself did not permanently enter into reaction, we had hitherto found to have no influence on the rate of reaction. For example, the accumulation of carbon dioxide does not accelerate the rate of oxidation of carbon monoxide, nor of methane, etc.; nor does nitrogen enhance the rate of decomposition of ammonia. The function of nitrogen in catalyzing the polymerization of acetylene is to furnish additional clustering and polymerizing centers in the form of N2+ ions. We have just found that the number of acetylene molecules condensing for each N2+ ion is approximately 20, the same as for each C2H2+, and that polymerization by both routes proceeds simultaneously. In other words