II] MODEL OF A MONOPODIUM 15 



such signs as the position of the bracts, the relative lengths 

 of the various axes, and the degree of advancement of the 

 flowers in different regions. 



There are, however, certain specially difficult cases of 

 mixed inflorescences, in part racemose i.e. monopodial 

 and acropetal in part cymose i.e. sympodial and definite. 

 Such of these as concern the plants here treated will be 

 dealt with in the sequel. The same may also be said of 

 some peculiar forms not readily brought under either 

 head without a special study of the development of the 

 whole inflorescence. 



The principal point for the moment, and it is a 

 fundamental one, is the difference between the uniaxial 

 or monopodial, and the multiaxial or sympodial type of 

 inflorescence in the sense referred to above. 



A rough illustration of the principal differences between 

 the two types may be furnished by supposing a ball of 

 string, from which we slowly draw out one end : as the string 

 is pulled out a short way, suppose we tie another short piece 

 of string in a knot just below the apex to represent a 

 bract and flower-bud. We then pull the string out a little 

 further, slipping this first knot down as we do so, and tie 

 another knot round beneath the apex, and so on. The 

 final effect will be that of an axis, with the knots developed 

 in acropetal succession, as it were, down its length. 



In this way we should produce an approximate model 

 of a monopodium, in so far as the axis is one length of 

 string. 



Now suppose we start by pulling the string out a short 

 way and then stop ; we then take another piece of string 

 of about equal length, and tie it just beneath the upper 

 end of the first piece " hanging " it, as it were. Mean- 

 while suppose a flower-bud to have formed at the end of 

 the strangled piece, representing the head, and a bract to 



