REVISION OF THE KING SNAKES. 11 



that is a constant for the genus, but in the kinds of formulae pos- 

 sessed by any single form or group of forms. No individual ever 

 drops more than three rows on a side, and only certain formulae of 

 this kind occur. Thus while any individual that possesses a maxi- 

 mum of 25 rows may drop three rows on each side, making the formula 

 25-23-21-19, in only two forms (multicincta and pyrrhomelaena) is it 

 common for an individual with a maximum of 23 rows to have 17 

 rows at the posterior end, and in only three others has such a situa- 

 tion been observed to occur even rarely (as in polyzona, microjyholis, 

 and one specimen of iriangulum) . All others having a maximum of 

 23 rows possess the formidae 23-21-19, 23-21, or 21-23-21-19, or 

 21-23-21; those having a maximum of 21 rows never have a mini- 

 mum lower than 17; 15 rows is reached ohly by those that never 

 have more than 19. 



It is very noticeable that some forms in the genus (elapsoides in 

 particular) are undergoing, or have undergone, a reduction in number 

 of scale rows concomitant with reduction in bodily size and in scutel- 

 lation. This shov.s that evolution in Lampropdtis may involve 

 reduction in scutellation, and that this reduction may be one evidence 

 that evolution away from the primitive type has taken place. On 

 the other hand, there is evidence from at least one instance (-floridana) 

 that evolution has taken place by increase in scutellation, including 

 increase in number of scale rows. From these definite instances of 

 evolution, involving change in number of scale rows, it is shown that 

 the formulae may be represented in descending order as follows : 



21-19 

 21-19-17 

 19-21-19-17 

 19-17 

 17-19-17 

 17-19-17-15 

 17-15 

 15-17-15 



In some forms, between the formulae 21-19 and 21-19-17, there 

 occurs the formula 19-21-19, and then 21-19-17 may occur but 

 rarely or not at all (the getulus group in particular). Reduction 

 takes place by a shortening of the rows to be lost, proceeding first 

 from behind forward and later, in addition, by a drop in the region 

 behind the neck, approximately opposite to the twentieth to thir- 

 tieth ventral scute; from this point the loss proceeds in both direc- 

 tions. Exactly a reverse in this process is the means by which a 

 higher form.ula is attained, and is best illustrated hj floridana. The 

 added row enters at some point anterior to the middle of the body, 

 and extends its length from here in both directions. 

 186550— 21— Bull. 114 2 



