280 



SCIENCE 



[N. S. Vol. L. No. 1290 



that I venture to give a few typical examples 

 of the information which I have been gather- 

 ing. I may recall that the observations are 

 made in a very large, dark, damp room, semi- 

 subterranean, in which the temperature varia- 

 tions on the thermograph rarely reach 1° per 

 day. A compensated ball m of about .6 gram, 

 movable on a quartz fiber, is gravitationally 

 attracted by an external kilogram, M, in the 



The readings were now averaged per semi- 

 day (a.ai. and p.m.) and also per day and an 

 example of results is given in Fig. 2, Ay show- 

 ing the double amplitude or total excursions 

 of the aperiodic needle, resulting from the 

 combined attraction of gravitational and 

 radiant forces. Ay varies from much below 

 3 to much above 7, even in the short interval 

 between July 15 and August 4, and in spito 



M S5 cA>^^ 



Fig. 1. 



given environment. When M is on the right, 

 or on the slightly warmer side of the region, 

 m is deflected toward larger numbers y, of the 

 scale. In the other ease, the reading 2/ is in 

 smaller figures. The actual displacement of 

 m is x^ .1455 y in apparatus I; and x = 

 .0214 y in apparatus II. The needle carrying 

 m (balanced by an opposite m) pointed north- 

 south, the case being on this side of the large 

 central laboratory pier. The basement wall 

 is over 4 meters off on the east, and over 30 

 inches thick. This therefore receives solar 

 radiation, directly or indirectly, in the morn- 

 ing. 



Fig. 1, for instance, shows the direct read- 

 ing, y of apparatus I, on July 31 and August 

 1, the mass M being passed alternately form 

 east (larger reading) to west, at intervals of 

 about an hour. 



of the fact that the laboratory temperature is 

 practically constant. It follows therefore that 

 the ball 1/ and the east wall reciprocate in 

 their radiant exchanges, almost directly, in a 

 way of which the room temperature gives no 

 interpretable account. 



On the top of Fig. 2 I have inserted the 

 atmospheric temperatures kindly furnished by 

 Meteorologist Charles S. Wood, of the Provi- 

 dence Station, U. S. W. B. It is obvious at a 

 glance that the two groups of curves, those 

 for Ay and that for external atmospheric tem- 

 perature are of the same kind; but the Ay 

 curves follow the temperature curve with a lag 

 of one to three days. Whatever radiation 

 falls on the outside of the east wall of the 

 laboratory, shines in a subdued form on the 

 ball M a few days later. In the dark room 



