440 TF". LeConte Stevens — Experimental 



through the series, this shows that the formula gives results 



that are progressively too high or too low, as the case may be. 

 The explanation just given applies equally to Tables II, III 



and IY, the last two of which represent a fall of temperature 



through 80°. 



Table I. 



R 9^ 29 m 9 h 33 m 9 h 39 m 9 h 45 ,n 9 h 53"' 10 h 3™ 10 h 19™ 10 u 38 ra ll 1 ' 18 m 



T 334-16 329-20 324 26 319-33 314-43 309-52 304 61 299-65 294-72 



T 289-16 289 20 289"26 289-33 289-43 289-52 28961 289 65 289"72 



T-T 45 40 35 30 25 20 15 10 5' 



E 419-1 366-9 315-4 264"6 218-6 171-0 126-6 83"8 41-3 



W 0-403 0-352 0-302 0-255 0-210 0-165 0*122 0-080 0-040 



S 0-783 0-680 0-578 0-485 0-392 0-306 0-225 0-145 0-069 



W-f-E 962 959 957 964 960 965 964 966 968 



S-s-E 187 185 183 183 179 179 178 175 167 



Table II. 



H 9M7 m 9 h 53 m 9 h 5S m 10 h 5 m 10 h 13 m 10 h 24 m 10 h 38 m 10 h 59"' ll h 37 m 



T 332-98 328-08 323-18 318 32 313-46 308-63 303-S2 299-00 294-13 



T 287-98 28S-08 288-18 288-32 288-46 288-63 28882 289-00 289-13 



T — T 45 40 35 30 25 20 15 10 5 



E 426-1 374-4 321-4 271'0 2219 173-7 127-3 84-7 42"0 



W 0-403 0-352 0'302 0"255 0210 0-155 0122 0-080 0'040 



S 0-783 0-680 0-578 0'485 0-392 0-306 225 0-145 0'069 



W-f-E 94 6 940 940 941 946 950 95*8 945 952 



S-s-E 1S4 182 180 179 177 176 177 171 164 



Table III. 



H 10 h r6"' 10 k 20 m 10 h 25 m 10 h 30 m 10 h 36"" 10 h 44"' 10 h 54 m ll h 9 m ll h 32 m 



T 384-21 374-25 364-33 354-40 344'49 334-57 324-65 314-70 30470 



To 289-21 289-25 289'33 2S9-40 2S9-49 289-57 289-65 289"70 289-70 



T— T 95 85 75 65 55 45 35 25 15 



E 347-6 3027 2563 214-2 175-0 138-7 103 9 72 2 42'0 



W 0-999 0-865 0-739 0-619 0-508 0403 0302 0210 0122 



S 2-113 1-803 1-514 1-249 1006 780 0-578 0\392 0-225 



W-f-E 287 286 288 289 290 290 291 291 290 



S-S-E 608 596 591 583 575 562 556 543 536 



Table IV. 



H 10 h 35 m 10 h 39 m 10 h 43 m 10 1 ' 45 m 10 h 54 m ll h 2 m ll h 12 m ll h 27 m ll h 49 m 



T 



385-27 



37530 



365-35 



355-43 



345-50 



335-61 



32568 



315-71 



305-81 



T„ 



290-27 



290-30 



290-35 



29043 



29050 



29061 



290-68 



290-71 



290-81 



T-T 



95 



85 



75 



65 



55 



45 



35 



25 



15 



E 



432-0 



376-0 



320-0 



266-2 



218-2 



1730 



129-5 



89-8 



526 



W 



0999 



0-865 



0-739 



0-619 



0-508 



0-403 



0-302 



0-210 



0-122 



S 



2 113 



1-803 



1-514 



1-249 



1-006 



0780 



0-578 



0-392 



0-225 



W-r-E 



231 



230 



231 



233 



233 



233 



233 



234 



232 



S-s-E 



489 



480 



473 



469 



461 



451 



446 



437 



428 



If now each one of the series of numbers in column ~W -f- E 

 or S -f- E be divided by the mean of the series the successive 

 quotients will differ but little from unity. The corresponding 

 horizontal column of results we may call the column of devia- 

 tions. If the numbers in this column be represented by a curve, 

 with temperature-differences for abscissas, this curve should be 



