"If an ordinary thermometer tu])e could V)e magnified, the irrcfrii- 

 larities of the hore would look sometiiing like the varying diameter of 

 the temperature column in Fig. 1. On the right side of the tul)e 

 there will be noticed the ordinary scale readings, 10 degrees to 50 de- 

 grees, with the freezing point indicated. On tlie left of the tuhc have 

 been noted the real temperatures — not the heights of the mercury 

 column. The thermometer is assumed to be correct at 32 degrees, 

 but owing to the increasing and diminishing size of the inside of tiie 

 tube, the true temperature is 16 degrees when the thermometer siiows 

 '20 degrees,' 20 degrees when it shows '24 degrees,' 40 degrees wiien tlie 

 mercury stands at '37 degrees,' etc. There are .some instances where 

 so-called 'reliable' thermometers showed a divergence of 4 degrees ami 

 even G degrees from the correct temperature." 



The problem was rendered more difficult bj^ the inaccuracy 

 of instiiiments of all kinds manufactured since the World War. It 

 is the common experience of all persons desiring dependable instru- 

 ments to seek pre-war construction, but where skilled manufacture 

 was not in evidence, recourse had to be made to the matter of 

 careful tests of the available stock. 



NINETY PER CENT OF ORCHARD TEMPERATURES ARE 

 INCORRECT 

 The result of a year's tests of thermometers, which were 

 brouglit to the testing department by over one hundred fruit grow- 

 ers, gave interesting flg-ures. It was found that 43% of the ther- 

 mometers read too 

 high, 45% read too 

 low, 6% were struc- 

 turally out of order, 

 and only 6% were 

 accurate. (See Fig. 

 No. 2). , The two 

 hundred and ten 

 thermometers that 

 were submitted for 

 test by the orchard- 

 ists during the year 

 were not the or- 

 dinary kind, but in- 

 struments bearing 

 names of well- 

 known makers and 

 cost their owners on 

 an average of $5 

 each. One fruit- 

 grower* reported 

 "Ninety per cent of 

 the thermometers 

 we were using were 

 incorrect and after 

 having same tested 

 and marked, we 

 were able to save 



23 



Fig. 2 — These diagrams show the relative accu- 

 racy of a large number of thermometers which the 

 fruit growers of California submitted to the test- 

 ing bureau of the department of meteorology and 

 aeronautics of the Los Angeles Chamber of Com- 

 merce during the season of 1920-21. The circle 

 shows that 45% of the thermometers submitted 

 for test read too high; 45«Jf read too low; e% 

 were out of order and 6% were found to be ac- 

 curate. 



*Cal. Citrograph, Vol. 7, Pp. 72-78. 



