1908.] NATURAL SCIENCES OF PHILADELPHIA. 477 



those of these zones. The Northern American species found in zones 

 II and V are, in all cases, also found in III or IV, and hence can con- 

 ceivably have descended or ascended from one or the other of the 

 latter two. 



Considering all the Mexican and Central American species, etc., 

 occurring also (a) in South America, or (6) exclusively in South America 

 and the West Indies, or (c) exclusively in South America, the order 

 (according to the number of species) of the temperature zones in which 

 they appear is always the same, viz. : II, III, IV, I, V, the numbers 

 of species being respectively: (a) 91, 90, 60, 4, 1; (6) 29, 29, 18, 0, 0; 

 (c) 52, 51, 33, 4, 0. Zone II as a pathway for the extension of the 

 "South American element" has already been considered on page 474. 

 Some Anomalies in the Zonal Distribution. — Since Zone II is the only 

 zone continuously extending northward from South America, it is of 

 importance to notice, as Table 9 shows, that of the 70 Mexican and 

 Central American forms occurring elsewhere exclusively in South 

 America, 16, or 23%, have not been detected in zone II in our faunal 

 region. 



Of the 131 species, etc., common to Mexico, Guatemala and Costa 

 Rica, the great majority have been found in zone II at some point 

 (and hence presumably occur in it at other points), although they 

 may also inhabit other zones, but there are 28 exceptions. 



Sixteen of the species, etc., of zones III and IV of Costa Rica are 

 not found farther north, but 17 other species of the same zones and 

 country do so extend, although the Costa Rican areas of these zones 

 are discontinuous. 



In spite of the isolation of zone IV in Guatemala, no species of 

 Odonata are known to be restricted thereto, and species found there 

 and not known to occur in any other zone than IV, such as Cordule- 

 gaster godmani and Libellula foliata, are also found in Mexico and in 

 Costa Rica. 



Sixty-two and 27 species, etc., of zones higher than II found in Mexico 

 and Guatemala are not and are, respectively, found farther south, a 

 discontinuity of zones III-V existing south of Guatemala. 



Nine and 4 species, etc., of zones IV-V (but not lower) found in 

 Mexico are not and are, respectively, found farther south. 



Some conceivable explanations of these anomalies (suggested for 

 future investigation) are: incompleteness of data on the present dis- 

 tribution; that temperature does not limit the inhabitable area of 

 the species concerned, or that, limiting it, the species found in dis- 

 continuous parts of the same zone may at times, past or present, 

 31 



