1908.] NATURAL SCIENCES OP PHILADELPHIA. 479 



merits of rainfall show a much greater anmial variation^^ than in the 

 case of temperature, and a much longer series of observations is, 

 therefore, needed to determine the approximate average rainfall than 

 to ascertain the approximate mean annual temperature. Since such 

 series exist for very few places^* and the rainfall differs so much at 

 nearby localities (as the data gathered by the authors quoted 

 show), the endeavor to correlate the occurrence of certain species of 

 Odonata with rainfall differences appears to be premature,^^ although 

 it may be that precipitation has a more important influence on the dis- 

 tribution of these insects than has temperature.^^ The annual varia- 

 tion in rainfall, however, may bring about an annual variation in the 

 local Odonate fauna — a possibility which suggests that a proper under- 

 standing of the insect fauna may be obtainable only from the same 

 methods of accumulation of data as are practised by the meteorologists. 

 Mr. E. B. Williamson, as a result of his observations made while 

 collecting in Guatemala, has suggested in correspondence that "The 

 species [of Odonata] occurring at any location during the dry season 

 are those species of widest distribution, or, in other words, local species 

 are to be found in the height of the season. " In testing this sugges- 

 tion, the difficulty at once arises that we have no complete records of 

 the Odonata occmring both in the wet and dry seasons at the same 

 locality. An absolute essential for the study of this and other problems 

 connected with the seasonal distribution of these insects in our district 

 is a series of continuous observations for at least twelve consecutive 

 months in the same limited area. Under the present conditions the 

 best that can be done is to compare wet season captures at one point 

 with dry season collections at the nearest similar station. Thus 

 both Santa Lucia and Zapote lie on the Pacific slope of Guatemala, 



^ Escobar, Memor. Soc. Cien. "Antonio Alzate," XX, 1903 (see his figiires for 

 Mazatlan, e.g., I.e., p. 29). Harrington, Bull. PMlos. Soc. Washington, XIII, pp. 6, 

 19, 1S95. Sapper, Meteorol. Zeitschr., 1S92-1906. A still more recent review 

 of the distribution of rainfall in Central America is contained in Dr. Alfred 

 Merz's "Beitrage zur Ivlimatologie und Hydrographie Mittelamerikas " {Mittheil. 

 Vereins fiir Erdkunde zu Leipzig, 1906; 96 pp., 4 Beilagen; 1907, especially 

 pp. 9-23). An extended discussion of Dr. Merz's work is given in Meteorol. 

 Zeitschr., XXV, pp. 326 et seq., July, 1908. 



^^ These localities are mostly at the higher elevations, in the larger centres of 

 human population, wliile the majority of the species of Odonata are found at 

 lower levels. 



^^ See a note by Mr. Champion {Biol. Cent.-Amer. Neur., p. 53) and one by 

 Mr. C. H. T. Townsend {Ann. Mag. Nat. Hist., 6, XX, p. 289, 1897) on the 

 seasonal appearance of certain Odonata and Diptera, respectivelj', in our dis- 

 trict. 



^* Mr. F. M. Chapman has some interesting remarks on the influence of tem- 

 perature, independent of humiditj^ on the distribution of birds at Las Vigas and 

 Jalapa, Vera Cruz, Mexico {Bull. Amer. Mus. Nat. Hist., X, pp. 17 and 36). 



