230 



• KNOWLEDGE . 



[Oct. 12, 1883. 



this line appears single, but with the slit shut as closely as 

 possible, and the light from the sky very strong, when the 

 observer is possessed of good eyesight, it is just possible to 

 see that this D line is really double — the finest imaginable 

 line of light appearing exactly in the middle of it, and 

 thus dividing it into two. 



□ Whenever there is moisture in the air which is 



likely to fall as rain, this D line (or lines) appears 

 thicker and blacker, and on the side towards the red 

 there is a shaded band from two or thrf e to five or 

 six times the breadth of the U line. This band has 

 been named the rain-band. If this band is very dark 

 and broad, and is visible when the spectroscope is 

 directed towards the sky, at an angle of -1.5" — that 

 is half-way between the horizon, and directly over- 

 head — then rain may be confidently expected in from 

 six to twelve hours. 



In Fig. 2 I have given an engraving of the D 

 Fig. 2. lines in the spectrum when there is no rain-band 

 present, and in Fig. 3 an engraving of the spectrum, 

 with a very strong rain-band showing in it such as I have 

 just described, as indicating that rain is imminent. The 

 two strong lines in the shading to the left of the D will 

 not be shown by any low-power spectroscope of the 

 ordinary construction. 



RAIN - BAND 



YELLOW 



It must be understood that this band may appear of any 

 degree of intensity, from even darker than it is shown in 

 Fig. 3, to the faintest sign of its presence in the spectrum. 



Some little practice is necessary to estimate the intensity 

 of the rain-band ; but this knowledge is generally soon 

 acquired. 



To fairly estimate the probability of rain and the time 

 at which it may be expected, it is necessary to observe 

 carefully the intensity of the rain-band at diflerent degrees 

 of elevation above the horizon. 



An hour or two after sunrise there is mostly a slight 

 rain-band visible near the horizon. This I should call a 

 Ion; faint rain-barid. It would be wrong to conclude that 

 this appearance will be necessarily followed Ijy rain. It is 

 a good plan to make rain-band observations about nine 

 o'clock in the morning, and to observe most particularly, 

 when the rain-band is seen, how high up in the sky it is 

 visible. 



In e&timating the intensity of the rain-band, Mr. Eand- 

 Capron gives a series of diagrams of the rain-land, six in 

 number, and refers all his observations to one of these, 

 estimating the variation in its intensity to about from 

 fifteen to twenty per cent. 



Mr. Corry, who has written some excellent letters to 

 Knowledge on this subject, assures me that after only a 

 few weeks' practice, he had no ditficulty in estimating its 

 intensity to ten per cent. 



I find for beginners it is sutficient to estimate the in- 

 tensity to twenty-five per cent, according to four degrees 

 of intensity, of which Fig. 3 may be considered as the 

 third, equalling seventy-five per cent. 



I cannot get diagrams engraved on wood of sufficient 



accuracy and delicacy to represent even these four degrees 

 of intensity. 



The diagrams given are, for wood-cuts, by far the best I 

 have seen, and do the engraver great credit; but even their 

 preparation to my satisfaction has caused a month's delay 

 in the appearance of these articles. 



I have referred first to the shaded band only in the 

 spectrum, because this only is shown in the simplest and 

 most inexpensive spectroscopes. 



Professor Piazzi-Smyth, who first suggested the use of 

 the spectroscope for predicting rain, used an exceedingly 

 powerful and costly instrument for the purpose, in which 

 the rain-band was shown as a nvmher of separate lines. 



LINES IN 

 RAIN:- BAND 



RilN-BAND. 



Thanks to the ingenuity of Mr. Grace, we are now pro- 

 vided with a portable spectroscope at a very muderate 

 expense, which will clearly show the two principal lines in 

 the rain-band. These two lines are shown in the diagram. 

 Fig. 4, and again, with the shading by which they are 

 always accompanied when they are strong, in Fig. 5, 

 and also in Fig. 3. The ordinary small spectroscope wiB 



V ! i ft 



Fig. G. 



not, I repeat, show these lines. Grace's spectroscope is 

 shown in Fig. G. Until I was able to supply this instru- 

 ment, a great number of my correspondents were not able 

 to see, let alone estimate, the intensity of the rain-band. 



In my next article I shall refer to the angle of disap- 

 pearance of the rain-band, the best method of making and 

 recording observations, and the practical conclusions to be 

 deduced from them. 



EPHEMEILA, OR DAY FLIES. 



THESE insects belong to the family which is scientifi- 

 cally called Ephemerida?. They are called day flies 

 on account of their short life, a single day sometimes 

 witnessing their entrance into a perfect state of development 

 and their death. They pass about two years in their larval 

 and pupal state. 



These insects are interesting and remarkable for a stage 

 of development which is very uncommon. When they 

 forsake the water where their larval and pupal state is 

 passed they creep out of the pupa case, and after resting 

 for a short period — from one to twenty hours — begin a 

 tremulous motion of their wings. Then they fly to the 



