To the Editor of the SCIENTIFIC AMERICAN: In your issue of April 15th, 1911, there appears an article on the production of artifcial rain for agricultural purposes, in which article certain rain producing apparatus is described, the invention of which is credited to a German whose name is given as Hartmann. Permit me to call attention to the fact that the system of rain producing described in your article, which I notice is credited to the German periodical Prometheus, is my invention, and that I have secured the following patents on it: Patents No_ Australia .....................................15,376 United States .......................958,937, 963,932 France...................................... 403,701 Belgium _____________________________________211,923 Argentine Republic, 5,684, 6,392, 6,642, 7,170, 7,171, 7,944, 7,949, 8,292 Uruguay ....................................412, 474 Including four German claims now pending in the Imperial Patent Ofce in Berlin. I write this letter in order that any wrong impression that may have been created by the article from Prometheus may be removed. New York city. Emilio OlssQn. The Diference Between Light and Heavy Winds To the Editor Q. the Scientific American: 1n the issue of May 20th, page 507, of the SOientific American, under heading, “Home-made Anemometer,” the formula to obtain true velocity of wind is B B P = 0.400 - S V'. Should n t this read P = 0.004 - S V'? 30 30 Would you please inform me just where you draw the line between light winds and heavy winds? Pittsfeld, Mass. P. N. Moore. [Our correspond e nt ,has very properly called attention to an unfortunate typographical error in the article referred to. The factor 0.400 was a misprint for 0.004. B We may add that the factor — (correction for 30 barometric prer,sure) can safely be neglected when the place of observation is near sea level. The choice between the larger and the smaller wind-plates must be decided by the observer's experience in the use of the instrument. It will depend, of course, upon the strength of the spring and the length of the scale, a well as upon the force of the wind. The whole subject of normal pressure-plate aneIv;eters, and the various forms of such apparatus (ihe earliest of which dates back to the year 1724) will be found fuHy discussed in C. Abbe's “Treatise on MeteoroJ.ogical Apparatus and Methods,” which was published as an appendix to the Annual Report of the Chief Signal Ofcer for 1887. I is hardly necessary to state that the rough-and-ready .instrument described by Mr. Gilbert should, whenever possible, be compared with a standard ane mometer-or, more conveniently, a Richard anemo--cinemograph, which shows the velocity of the wind directly, at a given moment-and its error thus determined. Having obtained a fairly acurate instrument, :t should be further borne in mind that the velocity of the wind near the earth's surface is nearly always very much less than that prevailing at the altitude of an Qrdinary aeroplane fight; and that the existing statistics of the varaton of winds with altitude are mere averages, subject to wide fuctuation. On the whole, the anemometer is, as yet, of but moderate utility in aeronautics,-Ev.] Self-luminous Night Haze UNDER the above title Prof. E. E. Barnard, of the Yerkes Observatory, publishes in the Proceedings Of the American Philosophical Society an important contribution to the subject of nocturnal sky-light. His prper is especi8l1y interesting for the reason that, at the time of writing, the author appears to have been unacquainted with the publications of Yntema, Maurer, and other Europeans on the so-called “earth-light,” a brief account of which was published. in the Scientific Ameican, Nov. 19th, 1910, p. 394. Barnard's observatiQns confrm and supplement those of his European confreres. On moonless nights the sky is often more or less distinctly luminous, and this luminOSity cannot be wholly attributed to the difusion of the general starlight by moisture in the air. At times the illumination is so great that the face of a ordinary watch can b read by it. There seems to be no doubt that this luminosity is of an auroral nature, as it has been shown that the spectrum of the aurora is essentially always present on a clear dark night. In describing this phenomenon the author merely corroborates the observations of several earlier writers, but he proceeds to describe a particular manifestation of this luminOSity-or possibly a phenomenon distinct th'refrom-to which attention does not appear t have been previously directed. This cQnsists of strips and patches of luminous. haze, which have been observed at Yerkes Observatory severa.] times during the past year. It is not oonfned to any particular region of the sky nor to any hour of the night. It always has a slow drifting motion among the stars, comparable to that of ordinary hazy streaky clouds that are often seen in the daytime The streaks are usually straight and difused, and as much as 50 degrees or more in length, a n d 3 degrees to 4 degrees or more in width. In some cases they are as bright, or nearly so, as the average portions of the Milky Way; i. e., they are decidedly noticeable when one's attention is called to them. They are apparently ahout as transparent as ordinary haze. Sometimes, when seen near the hori?;on, where they may be rather broad, they have strongly suggested the “dawn” or glow that precedes a bright moonrise_ Their luminosity is uniformly steady. The writer believes this haze to b something quite diferent from the noctilucent clouds that were seen for several years follOwing the eruption of Krakatoa, but only during the short nights of summer; whereas the haze has been seen through last autumn and winter. The noctilucent clouds, with which Jesse's name is always associated on account of his indefatigable study of them, were pretty conclusively shown to consist of fne volcanic dust, foating at so great an altitude (upwards of 50 miles) as to receive and refect the light of the sun long before and after the ordinary clouds. Prof. Barnard believes what he has seen to be ordinary haze, in some manner rendered seLf-luminous. In one instance he watched strips of this luminous substance from before dawn until, as the daylight killed their luminosity, they were seen to be strips of ordinary haze. Meteorologists who read Prof. Barnard's paper wiH undoubtedly be reminded of the references that are Qccasionally made by students of clouds to “auroral cirrus;” for although the author throughout his p8Jper speaks of “haze,” he is evidently referring to cirrus or cirrostratus clouds-which are quite distinct from haze in the usual meteorological sense. Cirrus seen in the daytime often simUlates the forms Qf the aurora, and that there may be some connection between the ,two phenomena has often been suggested. A Lake with a Roof THE great salt lake at Obdorsk is nine miles wide and seventeen miles long, yet except in a few places it is solidly roofed over with a deposit of salt which is becoming thicker and thicker each year. AbQut the middle of the last century salt erystals frst began to gather upon the surface of the water. Year by year, owing to the evaporation of the water, the crystaJ.s became more numerous, and then caked together until this great roof was formed. In 1878 the water beneath this salt-crystal roof found an underground outlet into the Obi River. This :owered the lake's surface about three feet leaving that distance between the water and the roof, and each year this distance has been diminished by the constant addition of salt crystals to the roof. Many s.rings surround this lake. Their water fows over the roof and evaporates there, and thus continually adds to its thickn(,ss. After many years the springs will probably heeome choked with their own deposits, and then the whole wi become covered with earth, so that a great salt mine will be formed-a treasure for the Sibrians hundreds of years to come.