the belly so that when we have the specimen fairly on its 

 back we can only just see part of the last row of the other 

 scales (Fig. 4). These shields are called ventrals and they 

 extend from the posterior sublingual shield of the head all 

 along the body to the anus or vent. Beyond this we are 

 dealinor with the snake's tail and the ventrals are now called 

 sub-caiidals (Fig. 5). 



The last ventral shield is called the anal shield (Fig. 5). 

 It is generally double and so are the sub-caudals. The latter 

 are divided down the centre by a zigzag line. 



The number of ventrals and sub-caudals, though very 

 variable within limits, is most important ; for it is one of the 

 ways in which we differentiate snakes. 



Lastly, we will put the snake on its belly again (Fig. 6) 

 and see that the rest of its body is covered with small 

 scales. 



These scales are arranged in regular rows, varying in 

 most snakes from about 13 to 25 longitudinal rows. The 

 scales are generally all of the same size. It is most import- 

 ant to ascertain the number of rows of scales tha^ a snake 

 has, and we do this by counting the rows transversely. 



The most convenient way to do this is to start at the 

 extremity of a certain ventral and follow the line of scales 

 right over the body to the extremity of another ventral. 

 Notice that the row does not go over exactly at right angles 

 to the body but slants considerably (Fig. 6), 



The rows should be counted at a distance from the 

 head equal to about a quarter of the length of the body. 



The number of rows is practically always odd, so that 

 there is a sinofle row down the middle of the back called the 

 vertebral row (Fig. 6). (Don't confuse this with the ventral 

 shields.) 



Besides the number of rows there are two or three other 

 points to be noticed about the scales. First, are they all of 



