PATTERN OF BUDDING 333 



I call lateral groups. Whether he was also referring to the series (4) arising from the bases of the lateral 

 groups is not clear. 



The development of group 2 and the subsequent lateral tripartite groups can best be seen in the 

 aboral region of the main zone by examining cormidium I of a young specimen (PI. XIV, figs. 2-3). This 

 region looks very complicated in older stages (PI. XIV, fig. 4) ; when mature it is almost impossible 

 to analyse this first cormidium without knowledge of the younger stages. Cormidium I has room to 

 expand because it lies at the aboral end of the main series, consequently the primary and lateral groups, 

 especially the first, are better developed and more widely separated than those of the other cormidia. 

 Its aboral end forms a projection (PI. XIV, fig. 3) on which the youngest groups can be seen. Two 



GZ OF GON, 



u 2 mm 



Text-fig. 15. Physalia physalis. Cormidium 5 (oral zone) of a young specimen number 11, less group 1, to show the three 

 parts, A, B, C, of an 'Urgruppe'. Note the endodermal villi. GZ of GON. z = gonozooid mentioned on p. 347. 



XII. 



beautiful young specimens, measuring 2-9 and 8 cm. in float-length, show clearly that at this stage 

 there are at least eight lateral groups in cormidium I. They decrease in size according to youth, so that 

 the one at the aboral end is a very small bud (PL XIV, fig. 3). Each group consists of a gastrozooid, 

 a gonodendron and a tentacle. They look as if they might be, and indeed at one time I thought they 

 were, beginnings of new cormidia, but in each of the other cormidia of the main zone there is a 

 counterpart of this series of lateral groups. As previously stated, there is evidence that the latest 

 (youngest) cormidia to appear do so at the oral end of the budding-zone, whereas in the lateral groups 

 the pattern of development is reversed, the youngest groups appearing at the aboral end of each 

 cormidium. When fully developed, there may be from ten to twelve fan-shaped lateral groups (PL XV, 

 fig. 4) in each main zone cormidium. They bud successively (PL XVII, fig. 4) each from the base of 

 its predecessor and as growth proceeds they curl round (PL XV, fig. 6) so that the last small lateral of 

 one cormidium can be seen lying to the oral side of the first and largest tentacle of the cormidium that 

 lies at its aboral end. 



(4) Secondary basal branches. These tend to occur wherever there is room for further growth and 

 expansion, that is, on the leeward side of the float, at the aboral ends of the cormidia, and, in the case 



