Greenhouse Notes. 169 



averages throughout the test as shown (on page 170) in a majority of 

 oases when no pressure was present. 



Av. temp, at boiler with 1 lb or more pressure 213 2-15 



" " of thermometer No. 2, Riser C, 1 lb or more pressure 214 7-15 



« " " " No. 3, Riser C, " " 202 11-15 



« " " " No. 4, Riser B, " " 212 8-15 



« " " " No. 5, Riser A, " " 214 2-5 



« " " " No.6,UpperreturnB/' " 205 3-5 



« « " " No. 7,Lower return C," " 162 3-5 



These figures fully agree with the results obtained in the tests of the 

 previous winter, and demonstrate the utility of pressure in forcing steam 

 through long and unfavorable circuits. The observations then made 

 as to the effect of different rates of pressure on the temperature within 

 the pipes were as follows: 



1 lb of pressure increases the temperature 1 — 3° 



2 lbs " " " 3—4° 



3 lbs " « « 5—8° 



5 lbs " « " 12—14° 



These figures only roughly approximate the temperature under dif- 

 ferent rates of pressure, and the ratio is not wholly a constant one in 

 practice, but in a general way about the same ratio holds good in the 

 present tests. 



The average temperature of the thermometer in the lower return is 

 far below that of the others as well under pressure as without, and this 

 difference was more noticeable towards the last of the test. This serves 

 as a good illustration of the bad effect of uniting returns before they 

 are carried below the water level of the boiler. The water level of the 

 boiler reached about to the point where the upper return dropped into 

 the oblique 2-inch pipe of the lower one, and doubtless whenever the 

 water was low enough to admit of an air connection here, the steam 

 returning from the upper one would serve as a trap to shut off the cir- 

 culation from the lower, a fact which was frequently demonstrated by 

 closing the valve of the upper return, when a circulation would at once 

 develop in the lower one, with a very sudden rise in the temperature of 

 that thermometer. In the latter part of the season the water was kept at 

 a lower level to provide more space for the formation of steam in the 

 boiler, and this difficulty was more noticeable. 



Influence of crooks and angles. — As previously explained (page 165), 

 a section was cut from the middle riser, B, Jan. 31, and re-connected 2 



22 



