SOLAR ECLIPSE OF JANUARY 3, 1908— ABBOT 



35 



Apparatus 



•\ concave mirror of 50 centimeters diameter and only 100 centi- 

 aneters focus, mounted equatorially and driven by a clock served to 

 produce a very intense image of the corona.^ A small gmdmg 

 Lescope was attached to the mirror frame, so that the observe 

 might point the mirror toward any desired object. In the focus of 

 the mirror was placed the bolometer. A glass plate three milhmeters 

 thick was fixed close to the bolometer, between it and the mirror, so 

 that the radiation examined was thereby limited to wave-lengths 

 less than about 3^. This device prevented any exchange of rays ot 

 lono- wave-lengths between the bolometer and the sky, such as pro- 

 duced negative deflections when the bolometer was exposed toward 

 the corona in 1900.^ The bolometer had blackened platmum strips 

 8 millimeters long and 0.7 millimeter wide and of 0.5 ohm resistance. 

 \ metal diaphragm with circular aperture of i millimeter diameter 

 was fixed between the glass plate and the central bolometer strip, so 

 as to limit the region of the corona examined at each observation to 

 an angular area of about 3' of arc in diameter. 



\bout 10 centimeters in front of the bolometer was a self-closing 

 blackened metal shutter which cut off the beam excepting when 

 designedlv opened. The opening of this shutter therefore exposed 

 the central part of the bolometer to such rays as are transmissible 

 bv fflass Between the shutter and the glass plate, and close to the 

 latter was a special screen composed of a thin stratum of asphaltum 

 varnish laid on one side^ of a plane parallel glass plate 3 milhmeters 

 thick This screen was held out of the beam by a spring, except 

 when designedly interposed- Its property, when used, was to cut ott 

 nearly alfthe visible part of the radiation, while transmitting nearly 

 all of the infra-red ravs transmissible by glass. The transmissibility 

 of this screen for rays of different wave-lengths follows: 



Wave-length 



Transmissibility. • 



0.50 



0.55 



0.60 

 0.04 



0.65 ' 0.70 I o 80 i.oo I 1.20 1.60 2.00 



0.10 1 0.20 0.41 0.66 080 0.90 092 



'The mirror was freshly silvered and polished on the day before the total 



^'^^^Negative deflections in those experiments were due to the fact that the 

 card screen used was warmer than the effective temperature of the sky^ not 

 as Deslandres intimated, because any kind of rays cools rather than warms 

 when absorbed. 



' The side nearest the bolometer. 



