sun, I do not know. I saw them three times, and attempted, at the last moment, to get another observation, but at the critical moment, a little cloud passed over the sun, and I hastened to observe again the sun for the third contact and attending phenomena. At each of the observations, by careful comparison, they appeared exactly of the same magnitude, and both as red as Mars. I looked closely for twinkling, but they were as free from it as the planet Saturn. They both, at the time, seemed to my eye and mind to have a small round disk about like the planet Uranus. Whether the disks were imaginary or real, I cannot tell, but every time I saw them (the stars) the disks attracted my attention. Immediately after totality I recorded the following in my note-book. "Saw two stars about 3° southwest of sun, apparently of the fifth magnitude, some 12' apart, pointing toward the sun. Both red." Last evening I experimented with my telescope with the same aperture and eyepiece, to verify or to change my estimate as to distance apart, and I find that they appeared a little farther apart than half the distance between Mizar and Alcor, or, say, about 7' instead of 10' or 12' as estimated at the time. Though at the time I estimated the direction of the stars as southwest of the sun, I am inclined by reflection to change it to south of west. One of the stars may have been Theta Cancri, and yet there are several chances against a planet being exactly of the same magnitude, color and appearance, as a star; and it appears to me that Theta is most too far north. I would not be surprised, if the truth were known, if both were planets. Prof. Watson, who claims to have seen a new star 24° southwest of the sun and of the 45 magnitude, may, when he publishes his observations, throw a little light on the subject. Should he have seen the same, of course the priority of discovery belongs to him by about two minutes. That one or both are intra-Mercurial planets, I have no doubt whatever. It may not be out of place here to say that I was led to the discovery by an accident, the result of carelessness. A strong and fitful breeze was blowing from the southeast which shook my telescope, the equatorial mounting being not very substantial, part of it having been made after reaching Denver. To prevent this shaking, I laid a pole about ten feet in length across the tube near the eye-end, the other resting on the ground to the west of the telescope. Then, for fear it would fall against my head, I tied it to the tube with a string. As I followed the sun, the pole would easily drag along the ground. I intended to remove the pole at the commencement of totality, but in the excitement I forgot it. When I began to sweep along the ecliptic for Vulcan, I found to my horror, that if I attempted to move the telescope to the east of the sun, the lower end, by plunging into the ground, would prevent me, and I was therefore compelled to devote all my time to the region west of the sun, which accounts for my getting so many views of the objects. From this cause, also, the sweeps were very irregular. I wish astronomers with large instruments would make the attempt to observe Theta Cancri, before the sun withdraws too. far, and if successful then the objects seen by me, we need not despair of their being again observed at some future time without a total eclipse or even in transit. Let the experiment of detecting a star of the 45 or 5 magnitude, and from 2° to 6° from the sun, at a large or annular eclipse, be tried. The result may be to settle this question, and remove all doubt. Respectfully yours, Rear Admiral JOHN RODGERS, LEWIS SWIFT. Superintendent United States Naval Observatory, Washington, D. C. SCIENTIFIC INTELLIGENCE. I. CHEMISTRY AND PHYSICS. 1. On the Determination of high Melting and Boiling Points.CARNELLEY has improved and extended the method for determining high melting points which he proposed in 1876. The method is the usual one for specific heat, the temperature being calculated from the ordinary formula. Two sources of error have now been detected and corrected: one, the excess of heat coming from the suspending wire of the crucible which proved to be large enough to influence the result; and the other, a false water equivalent for the calorimeter. The former error was eliminated by using a finer suspending wire; the latter, by calculating the true value from the formula MS(T―0)=(W+w) (0−t), in which M the weight of the crucible, S the specific heat of platinum, T the melting point of a salt as determined by a thermometer, t the initial, the final temperature, W the weight of water and w the water value of the calorimeter, are all known but the last, several salts whose melting points were accurately known being used in the experiments. In this way 98 was fixed as the calorimeter constant. The values obtained after making these corrections, gave melting points entirely trustworthy. The numbers representing these are given in a table, varying from boric acid B(OH), 186°, to sodium sulphate Na,SO, 861°, and including about a hundred salts. In connection with WILLIAMS, Carnelley has applied the above results to the determination of high boiling points. Fragments of salts whose melting points are known are placed in the vapor In and examined to see if they melt. Various devices are employed in applying the method, which are described in the paper. the case of HgCl, for example, NaCIO, melts in its vapor while NaNO, does not; hence its boiling point is between 302° and 316°. Observation with a thermometer gives it as 303°. Anthracene vapor fuses KNO, but not KCIO,, hence its boiling point is between 339° and 359°. Thallous iodide TII, boiling, melts NaCl and Pb(PO), but not PbP,O, or Na,CO,; its boiling point is therefore between 800° and 806°. The authors hope shortly to be able to fix the boiling points of K, Na, Mg, Tl, etc., in this way. Jour. Chem. Soc., xxxiii, 273, 281, July, 1878. G. F. B. 2. On the Vapor Density of Thallous chloride and Lead chloride.-RoscoE has simplified Deville and Troost's method of vapor density, desiring only to give it sufficient accuracy to fix molecular weights. A long-necked porcelain globe, of about 300 c.c. capacity, containing from 3 to 9 grams of the substance to be examined, is loosely closed with a stopper of burned clay and placed in a muffle until no more vapors escape and the temperature is constant. The globe is then removed, allowed to cool and the amount of substance it contains is determined by analysis. The temperature of the muffle is fixed calorimetrically by placing in it at the same time a weight of platinum of known value. It may be checked by the use of a second balloon containing mercury. In five consecutive experiments with mercury, the temperatures being 1019°, 894°, 815°, 972° and 1047°, the vapordensity of mercury vapor was found to be 6.92, 6·75, 6·91, 5·77, and 705; or 6'68 as a mean. In the case of thallous chloride seven experiments, at temperatures of 859°, 828°, 1015, 849°, 1026°, 852° and 837°, gave 8.15, 8-28, 8.06, 7-43, 8.75, 8.60 and 7.84 as the density of its vapor. The calculated value, assuming the molecular weight to be 238.07 and the formula TICI, is 8:49. The vapor density of lead chloride determined in this way, requiring however a higher temperature, was 9.12, 9.72, 9.51 and 9.64, at 1046°, 1089°, 1077° and 1070° respectively, the theoretical value for PbCl, with a molecular weight of 27714, being 9.62.- Ber. Berl. Chem. Ges., xi, 1196, June, 1878. G. F. B. 3. On the Action of Steam on ignited Charcoal.-LONG has made a series of experiments under the direction of Lothar Meyer, to ascertain whether the carbon dioxide produced by the action of steam on ignited wood charcoal stands in any fixed relation to the carbonous oxide. The charcoal was purified by treatment with boiling nitric and hydrochloric acids, and thorough washing, a porcelain tube 60 cm. long was filled with it, placed in a furnace, heated to redness and steam driven through it, the gas evolved being collected over water, and analyzed by Bunsen's method. In the first series of experiments, the results were not uniform for successive portions of gas, the CO, increasing and the CO diminishing. Moreover, two volumes of hydrogen should appear for every volume of CO, and one for each volume of CO; but this was not the case, the hydrogen being always too low. These results being confirmed by a second series of experiments, the author made direct tests to see if this error was due to an actual loss of hydrogen or to an introduction from without of carbon dioxide. On filling the tube with charcoal, displacing the air with hydrogen, closing one end, attaching a potash apparatus to the other and heating, 250 c. c. of CO, were collected in two hours. Hence the discrepancy in the analyses arises from absorbed gases in the charcoal, either CO, directly or free oxygen, and diminishes as the operation continues. Making allowance for this excess, no simple relation appears between the two oxides of carbon. The amount of carbonous oxide formed, however, appears to be determined by the amount of charcoal present, diminishing steadily as this lessens. This can only be explained by supposing that the first action of the steam upon the coal is to produce carbon dioxide, and that this, by the further action of the coal, is reduced in part to carbonous oxide, in precise analogy with the ordinary action of oxygen upon carbon. If the carbonous oxide be in contact with an excess of steam, a reduction of the steam takes place. CO+H,O=CO2+H2. In practice, probably all three of these reactions go on simultaneously, the relative quantities of steam and charcoal determining the proportion of the gaseous constituents.-Liebig's Ann., cxcii, 288, June, 1878. 2 G. F. B. 4. On the Reduction-product of Gum Elemi by Zinc-dust.CIAMICIAN has continued his reduction experiments with the gumresins and has now submitted gum elemi to the action of zincdust. Commercial gum elemi was treated first with cold alcohol, and the residue of this operation was recrystallized from hot alcohol, and obtained in long needles in wavellite-like groups. The reduction was effected as in the case of abietic acid, the vapors being passed over ignited zinc dust. From 800 grams of the crystallized gum 300 c. c. of a brown oil, lighter than water, was obtained. Distillation in steam gave a volatile portion (A) and a tarry portion (B). The former, after boiling with sodium, was fractionated. Two products resulted, one boiling at 111° and easily recognized as toluene; the other boiling at 158°-160°, a colorless oil of peculiar aromatic odor, was evidently ethylmethylbenzene. The first gave benzoic acid on oxidation, the latter a mixture of isophthalic and terephthalic acids. Hence both paraand meta-ethylmethylbenzene were present in this latter fraction. Portion B afforded after treatment, a colorless liquid, boiling between 250° and 252°, having an aromatic odor resembling that of naphthalene, and a specific gravity near that of water. Analysis and its oxidation products fixed it as ethylnapthalene. Hence toluene, meta-ethylmethylbenzene, para-ethylmethylbenzene, and ethyl-naphthalene are the products of the reduction of gum elemi by zinc dust.-Ber. Berl. Chem. Ges., xi, 1344, July, 1878. G. F. B. 5. On the Constitution of Starch.-MUSCULUS and GRUBER have made further experiments in support of the view advanced by the former of these chemists, that the saccharification of starch by diastase or dilute acids, should be considered a splitting into a dextrin and sugar simultaneously. They now find that the action of the above agents on starch produces: 1st, soluble starch, soluble in water at 50°-60°, colored wine-red by iodine when in solution, blue when dry, rotatory power [a]=+218, reducing power 6; 2d, erythrodextrin, soluble in cold water, colored red by iodine, both in solution and solid, easily attacked by diastase; 3d, achroödextrin a, not colored by iodine, rotatory power [a]+210, reducing power 12, partially saccharifiable by diastase; 4th, achroödextrin B, rotatory power [a]=+190, reducing power-12, unacted on by diastase in 24 hours; 5th, achroödextrin y, rotatory power [a]=+150, reducing power 28, unattackable by diastase, converted into glucose by sulphuric acid only after long boiling; 6th, maltose, formula C,2H22O11.H2O, rotatory power [a]=+150, reducing power 66, unacted on by diastase, but fermentable; 7th, glucose, C,H12O6.H2O, rotatory power [a] 57, reducing power 106, fermentable. The authors hence consider starch as (C12H20010), in which the exact value of n, which cannot be less than five or six, is yet to be determined. Under the influence of ferments or of dilute acids, it undergoes a series of hydrations and successive splittings, forming at each, maltose and a new dextrin having a less molecular weight. The value of n becomes less and less until the y form of achroödextrin is reached, which is transformed by simple hydration directly into maltose, which again splits into glucose.— Bull. Soc. Chim., II, xxx, 54, July, 1878. NH-· G. F. B. 6. Synthesis of Indigo-blue.-Early in the present year, BAEYER showed that oxindol was identical with orthoamidophenylacetic acid, and hence had the formula C.HCHCO. In order to produce isatin synthetically from this, it is only necessary to reverse the process by which oxindol is produced from isatin; i. e., to convert the CH, group in the above formula into CO. This the author has now succeeded in doing, not by direct but by indirect oxidation. From nitrosoöxindol CHCHNO)CO which contains the nitrosyl group in the right position, amidoöxindol C.H CH(NH,) CỔ is produced by reduction, and from this by oxidation with ferric or cupric chloride or even with nitrous acid, isatin CHCOCO {NH(NH,) CÓ NH CH, is produced. The reduction of this to indigo-blue, 4 NH effected long since by the author in conjunction with Emmerling, by the action of phosphorous chloride, acetyl chloride and phosphorus, he now finds to take place in two stages. In the first, produced by the action of phosphoric chloride upon isatin, an imidchloride of isatin results, CH, (COCCI N This, by reduction with phosphorus, zine dust, or preferably with ammonium sul -11 |