16 



year, while the tree measurements represent a great many years' 

 growth. The records obtained from the growth of plants in the 

 greenhouse invariably run higher than those obtained by the chem- 

 ical m.ethod or the tree measurements, but this is due to the fact 

 that greenhouse crops are grown only during certain seasons, when 

 the difference is often 30 percent, in favor of the morning light. 



There are m,any reasons why morning light should be better 

 than afternoon light. While we cannot here consider all the fac- 

 tors entering into a problem of this nature, two of special signifi- 

 cance in inhabited regions are smoke and dust. Although the 

 difference between m,orning and afternoon light may be greater 

 in com.mercial centers than in the country, our measurements of 

 trees 150 to 200 years old show that there was a difference before 

 the advent of civilization. According to our observations, made 

 from an elevated point, distant points are most clearly seen in 

 the early morning. As the day advances the atmosphere becomes 

 less clear, for fires are everywhere being made, the highways once 

 again become busy thoroughfares and a cloud of smoke and dust 

 is sent into the air to accumulate during the day and diminish 

 the light intensity. The atmosphere also seems to be more free 

 from clouds in the morning, but we have not attempted to study 

 sunshine records in regard to this point. Observations made at 

 high and low elevations throw some light on the effect of dust and 

 smoke on the air, the light at high points being superior, but the 

 difference gradually grows less during the day owing to the ascent 

 of smoke and dust. 



The foregoing observations have a practical bearing on the 

 direction in which the greenhouse should be placed in order to 

 obtain the best light. As already pointed out, a house running 

 in an easterly and westerly direction should be placed toward 

 the north to receive the superior morning light to the best advan- 

 tage. A house set 20 to 25 degrees north of east would be better 

 placed than one running directly east and west. 



DIFFERENT QUALITIES AND KINDS OF GLASS. 



As might be expected, the kind and quality of glass vary great- 

 ly. We made no attempt to test more than a few types, — chiefly 

 those used in greenhouses. Some of the samples were obtained 

 from the Boston Plate and Window Glass Company, a finn which 

 retails large quantities of greenhouse glass. 



The experiments were made out of doors, and the exposed 

 tubes were placed in boxes 2 1-2 inches from the glass which we 

 wished to test. The records shown in the table are averages of 

 four different experim.ents in which the exposures ranged from 2 

 to 2 1-2 hours. In all the tests 16 x 28 inch glass was used. No. 

 1 was double French, first quality; No. 2, double French, second 

 quality, and No. 3 American, third quality. The fluted glass was 



