GEOPHYSICAL LABORATORY. 87 



(3) By providing refrigerating- and drying plants of such a character that 

 the air within the building shall be maintained approximately as cool and 

 dry in summer as in winter. 



The last plan has proved successful in several recent structures, but could 

 not be considered in our case on account of the expense. The other two 

 are not satisfactory and, so far as can be judged by the experience of others, 

 are not worth the cost of installation. It was therefore thought wise to try 

 a plan which, so far as we are aware, has not been tried before, namely, 

 to place an insulating layer on the outside of the building. The exposed inner 

 surface is then an ordinary brick wall of ample strength, from which the 

 heat is kept away by an insulating layer of 6 inches of hollow terra-cotta, 

 with the air-spaces laid horizontal and closed at the ends. So far as can 

 be judged from the experience of the present summer, in which the heat 

 has not been excessive, this has proved successful. Furthermore, such an 

 insulating layer ought to help out the heating plant in winter, just as one is 

 kept warm by an overcoat. This outside layer is less expensive than brick 

 and structurally is almost as strong. 



The problem of installing a considerable quantity of heavy machinery in 

 the basement of a building without disturbing measuring apparatus of a 

 high order of refinement upon the floors above is a serious matter, which 

 has heretofore proved so nearly insurmountable that the usual practice is 

 now to take it out of laboratory buildings entirely and provide for it in a 

 separate building at a safe distance. This has two obvious disadvantages: 

 (i) The cost of the additional building, and (2) the loss of power in trans- 

 mission from the generating plant to the laboratory, which is considerable. 

 There appeared to be one possible way out, which has been adopted, but 

 the final result can not be comm.unicated until we have had further oppor- 

 tunity to study the behavior of mounted apparatus. The experience upon 

 which the scheme is based was gathered by Dr. G. F. Becker in a visit to 

 several of the European laboratories and observatories, undertaken during 

 the summer of 1903 and reported in the second Year Book of the Institution. 



His observation was, briefly, this : Only two of the large laboratories and 

 observatories of central Europe are satisfactorily free from mechanical dis- 

 turbances — the Cryogenic Laboratory at Leyden, Holland, and the several 

 observatories at Potsdam, Germany. The laboratory at Leyden is built upon 

 piles driven in a foundation of mud and soft sand, while the Potsdam 

 observatories are upon a sandhill which contains hardly a pebble of any size. 

 In both these laboratories the niercury mirror could be satisfactorily used, 

 and not elsewhere. From this observation it appeared that stability in the 

 sense required by laboratories was not obtained by seeking a rock foundation, 

 as has been the common practice. A modest and inexpensive attempt to 

 apply this observation in our laboratory was made in this way: The base- 

 ment floor of each room containing machinery consists of a block of Portland 



