A CATALOGUE OF OBSERVATIONS OF LUMINOtTS METEORS. 145 



3° below Polaris in three-qviarters of a second, leaving a bright streak for 

 three seconds. The meteor first appeared in the field of view, and passed out 

 of it. The spectrum of the early portion of the streak, behind the nucleus, 

 was a greyish diffuse baud |° in width. The spectrum of the nucleus was 

 appreciably the same. The brightest part of the streak, before it faded, was 

 brought into the field of view, weU situated parallel to the edges of the 

 prisms, and in the middle of the field for about two seconds. Its appearance 

 was that of a golden-yellow line of light about 5° iu length, some 4' in width, 

 tapering gently towards the ends, and perfectly sharp and well defined. It 

 was unaccompanied by any continuous spectrum, or any bands or other lines, 

 and it so disappeared from the ends towards the centre. 



" No. 15, 2^ 15™ A.M. Equal to a second-mag. star. Shot from ;u to a An- 

 dromedse in three-quarters of a second, leaving a streak for two seconds. 

 The train-spectrum was a diftuse greyish-white band, |° in width, and about 

 6° or 7° long, and faded away without any further change. The spectrum 

 of the nncleus showed prismatic colours. 



" No. 16, 2*^ 16" A.M. Equal to a second-mag. star. Shot from 6 Cassio- 

 peiae to /3 Andromedis iu half a second, leaving a streak for two seconds and 

 a half. The meteor was seen ■«'ith the unassisted eye. The last-fading por- 

 tion of the streak was examined in the spectroscope, where it appeared more 

 widely difi^used than when seen with the naked eye. Its colour in the spec- 

 troscope was a dull greyish white. 



" No. 17, 2'" 27"* A.M. Brighter than a first-mag. star. Shot from a Cas- 

 siopeise to o Honorum, leaving a streak for two seconds and a half. The 

 train-spectrum was a diftuse greyish-white band |° in width, not sensibly 

 brighter in any part, and it so faded. The spectrimi of the nucleus was 

 bright red and green. 



" Daylight at this time began to appear, and observations were obliged to 

 be discontinued ; the streaks of the August meteors might, however, already 

 be plainly divided into two classes. In the majority of cases, a bright yellow 

 line, having the unmistakeable appearance of the sodium line, was clearly 

 visible in the spectrum. In a smaller number of cases the spectrum was 

 merely a diff"use and greyish light band, or orcUnary spectrum of weak inten- 

 sity, resembling the spectrum of the glowworm's light. It will be interesting 

 to observe this form of meteoric spectrum, should it be more common among 

 the " phosphorescent " streaks of the November meteors than it was in August 

 last, when only five such purely "phosphorescent" streaks were noticed 

 entirely free from sodium light. 



" The spectra of the meteor-nuclei were seen in a few cases only Avith 

 distinctness, as they were in general overpowered by the brightness of the 

 sodium light whenever that was present. When the streaks were phospho- 

 rescent only, and free from sodium light, the nuclei in general presented 

 highly-coloured spectra, Hke the spectrum of solid matter at a glowing white 

 heat, or like the spectrum of an ordinary gas-flame containing white-hot 

 solid particles of carbon. A better night for observing nucleus-spectra would 

 be the 12th of December, when meteors leaving no trains are plentiful; and 

 they are for the most part very biilliant, radiating from some part of the 

 constellation Gemini. 



" That which spectral examination of the August meteors has most certainly 

 brought to light is the existence of an extraordinary quantity of the vapour 

 of sodium in their luminous streaks ; so that many of the streaks, especially 

 the most conspicuous and the most slowly-fading amongst them, consist of 

 nothing else but soda-flames for a great proportion (that is to say, the latter 



1866. L 



