HYDKOIDA. — BALE. 309 



other species is caused by the abrupt narrowing-in of the 

 leaflets at the base, so that openings are left along the sides 

 of the corbula. In the crucialis group and in Thecocarpus 

 these openings are guarded by the lateral spurs, which pri^ject 

 across them, and which support in the former a series of 

 sarcothecse, in the latter a hydrotheca. 



It is observed by Nutting that the closed corbula is often 

 erroneously described as having the distal edges only of the 

 leaflets bordered by sarcothecse, and he finds that in all cases 

 where he has dissected the corbulse the proximal edges also, 

 though overhpped by the distal edges of the contiguous 

 leaves, support a series of sarcothecse, which project into the 

 corbula. My experience (mainly of Australian species) is the 

 contrary of this. I have not found the condition above 

 described in any species, except very partially in A. calycifera. 

 The general rule is that when the leaves are united the distal 

 edge of one leaf overlaps the proximal edge of the next, and 

 is fringed with sarcothecse ; frequently the overlapping 

 portion is expanded into a free wing (often very large), which 

 then bears the marginal sarcothecse, and if the free extension 

 rises above the corbula, so as to have both edges free, then 

 both edges bear sarcothecse. Similarly, if the leaflets become 

 separate throughout, as in certain (presumably male) corbulse 

 of A. parvula, or separated at intervals only, wherever the 

 €dges are free sarcothecse may be developed. On the other 

 hand there are species — e.g., A. megalocarpa — which have the 

 margins of the component leaves united smoothly, without 

 any sarcothecse whatever, while a row of sarcothecse may run 

 up the middle of the leaf instead, and a secondary leaf or 

 crest may also spring from the same part. In some forms 

 many sarcothecse may be scattered irregularly over the 

 surface. 



The presence of a supernumerary leaf has often been noted, 

 but in some species more than one such leaf may be present, 

 and they may occur at the distal as well as the proximal 

 extremity of the corbula. In the closed corbulse of A. 

 parvula the first leaflet on one side gives origin to a secondary 

 free leaflet which runs along the side of the corbula parallel 

 with the rachis. The corbulse in general appear more prone 

 to abnormal developments than the trophosomes, and 

 irregularities of growth are, in some species, very common ; 

 for example, I have seen in more than one species a secondary 

 corbula given off laterally from the middle of the primary one, 



Aglaophenia divakicata (Busk). 

 Plumularia divaricata, Busk, Voy. " Rattlesnake," i., 1852, 

 p. 398. 



