r 



6 TirE AI!«ORPTION AND EiHSSION OF All!. 



waveleiiytlis uitli a plane grating and lens in vacuo. The six rods arranged star- 

 wise about the main cone are handles of a large nut (see s in Fig. 9) which serves 

 to loosen the outer cone in case the apparatus is to be taken down to renew the 

 f^rease or for any other reason. For the gi-easing causes the two cones to cling 

 together remarkal>ly. The screws, u' (Fig. 8), of the camera and the nut,/' (Fig. 

 9), of the slit-cairiage serve similar purposes. 



The entire spectrograph, with the exception of the main axes, the miciometer- 

 screws, and the slit-jaws is constructed of brass. The air-tightness of the instru- 

 ment depends no less upon tiie homogeneity of its material, especially of its 

 castings, than upon the accurate workmanship of its execution. 



All the closing surfaces of a chamber from which the air is to be exhausted, 

 no matter how accurately they may be ground together, require, in order to sustain 

 the vacuum, to be lined with a tightening medium. For this purpose, after having 

 tried various kinds of grease, I employ a mixture produced by melting together 

 five parts by weight of white wax with seven parts of the yellow vaseline of Chese- 

 Ijioush tt Co. of New York,' the mixture being twice filtered through a hot 

 funnel. Care nuist be taken to have it free from dust, since even small particles 

 of dust, especially filaments, if they get into the grease between the smeared sur- 

 faces, may endanger the tightness. In the heats of summer it is advisable to 

 diminish the proportion of vaseline. This grease is sensitive to light, and <>n ex- 

 posure shortly turns dark yellow; but this coloration does no harm in the present 

 case. I must wai'n experimenteis against the employment of white va.seline ; for 

 (at least, when mixed with white wax) it oxydizes tlie brass-\vork quite con- 

 siderably. The application of the coat of grease to the bearing surfaces is a matter 

 of great importance for the tightness of the apparatus, as well as for its efficiency. 

 Above all, before laying on the grease, the surfaces upon which it is to be laid 

 have to be cleaned with the most scrupulous care. Then, having carefully ex- 

 amined them to see that no little filaments have settled upon them, one should lose 

 no time in proceeding to the application of the grease. In doing this, I make use of 

 a spatula of a thin lancet shape made of instrument-makers' dogwood (Enonymus 

 eurojxeus). The gi-ease is to be laid on more or less thickly, according to the size of 

 the surface; but in any case it should be applied as uniformly as possible. In the 

 case of the great cones, I go all over the film so produced with my finger, which I 

 have previously scrupulously cleaned, rubbing the grease in, in two directions. Of 

 course, only one of two surfaces which are to be put togethei- is greased. 



When the suifaces ai-e small and are optically plane polished, such as those of 

 the fluorite window, the grease \vill better be laid on with the finest grade of 

 photographers' retouching pencil, so that the film may be as thin and narrow as 

 possil>le. If, in doing this, one keeps the line of gi-ease as far as jiossible from the 

 border of the inner ojiening, one will easily prevent any grease from getting into 

 the interior of the apparatus when one puts the window into place, a dry strip 



' In another part of this Report I have recommended watch-oil, — a discrepancy due to my only 

 recently having resorted to vaseline, and to the circumstance that this description of the instrument 

 has only been inserted in the report long after the main part of it was submitted. 



