Madreporarian Genus Turbinaria. 515 



occur between the polyps of Madrepora and of Turbinaria ; 

 but tlie nior])liological value of tlicsc differences has to be 

 c'stiniatiil by extended comjiarativc studies. We have yet to 

 iind out how far the livin;^ jioiyps are afi'ceted by the different 

 eonditions of life, due to the different forms of their coralla. 



Eighth Type of Groicth : Bifrontal. — Typically the Turbi- 

 narians carry j)oly|)s on only one side of the coralluni. Forms 

 occur, however, with polyps on both faces, this being brought 

 about l)y the fusion of two fronds back to back. Such fronds, 

 as might be expected, are always more or less erect. Among 

 the specimens contained in Mr. Saville Kent's collection there 

 exist complete series, showing the early cup-form and its 

 subsequent folding. The folds do not form open cylinders, 

 as in the mesenteriform method of growth, but thin vertical 

 plates by the opposing under surfaces fusing together. 



Similar fusions occur in all Turbinarians which form upright 

 fronds, but irregularly, whereas in the type of growth under 

 discussion they are the rule, and no free single fronds occur 

 except here and there as horizontal expansions round the base 

 of the corallum. These are to be considered as the continua- 

 tions of the original edge of the cup, i. e. of those portions of 

 the edge between the vertical folds. 



These bifrontal growths show the phenomena which I at 

 first took to be indications of regular periods of growth, but 

 which, as above stated, I now think are due to occasional 

 downward streamings of the ccenenchyma. 



These definite types cannot be supposed to exhaust the 

 possible transformations of the early cup. When our collec- 

 tions are more complete other normal methods of growth will 

 no doubt have to be added. 



Among the methods of growth shown by the specimens in 

 the Niitional Collection v.hich do not as yet admit of being- 

 ranked as ty})es there is one which 1 should like to describe 

 here, as it ajipears to be too definite to be accidental. Inas- 

 much, however, as beyond being slightly indicated in one 

 specimen, it actually occurs in only one other, it is not safe to 

 claim it as a type. 



One side of a conical cup is pulled down, as shown in 

 n. XX. fig. 1, which represents a specimen in the National 

 Collection. Two flaps, starting apparently in this way, 

 grow round the cup closely fused with its outside. On 

 meeting behind the cup they bend round again, and then 

 again, the foldings on each side being almost symmetrical. 

 Fig. 2 a is a diagram of the singular method of folding. 

 Fig. 2 gives a sketch of the specimen which has been 



