INFLUENCE OF TEMPERATURE ON DIFFUSION. 89 



the temperature in the South above the normal, followed by springs 

 during which the temperature was below the normal. The tem- 

 perature during December, 1902, was below the normal in the South- 

 west. (See Diagram II.) In January, 1903, it was above, but 

 below again hi February, and about normal or above in March and 

 April, the result being that only incipient outbreaks occurred in 

 northern Texas and probably South Carolina. (See Diagram II; 

 fig. 5, p. 20.) If the series of temperature maps (Diagrams I-V) 

 be compared with those showing the area covered by each invasion 

 the relation between abnormal temperatures and these invasions 

 will be clearly apparent. 



These records are those of the United States Weather Bureau and 

 are therefore correct so far as general field temperatures are involved. 

 When it comes to a consideration of the exact effects of temperature 

 and humidity upon the individual Toxoptera, however, the figures 

 will not apply with mathematical exactness, for the reason that to 

 secure this information it is necessary to learn the exact conditions 

 in the midst of the insects themselves at the exact time that such 

 data are being secured. To illustrate, the instruments of the 

 Weather Bureau kept in the shade may indicate a certain tempera- 

 ture, yet in a field perhaps a mile distant on a sunny day, and down 

 among the plants in the midst of the developing insects, there may 

 be several degrees difference in temperature. As will be noted 

 farther on, Mr. Luginbill has found this difference to amount in some 

 cases to several degrees. Besides, it is easy to conceive of other 

 conditions which might have precisely the reverse effect. Further- 

 more, there will be a difference in temperature as between fields with 

 a sandy and a clay soil or between a southern and a northern expo- 

 sure, or with a soil dry on the surface as against a soil with a wet sur- 

 face. It will be observed, therefore, that while the exact tem- 

 perature at which Toxoptera will reproduce, viviparously, is of 

 scientific interest, such information is of minor significance in the 

 field, where it is the more generally prevailing weather conditions, 

 such as are secured by the United States Weather Bureau, over wide 

 areas that become of greatest importance, Mr. R. A. Vickery, on 

 December 4, 1908, at Richmond, Ind., with 5 viviparous females 

 under observation, found that young were produced sparingly at a 

 temperature of 40° F. This was indoors, in a room slightly heated by 

 an oil stove so that the temperature was under control, and frequent 

 readings were made during the day. Under the same conditions 

 numerous young were produced when the temperature reached 45° 

 to 53° F. * 



