2 8 PENNATULIDA. 



ones. At the top of the shaft a short portion of the calcareous axis projects nakedly, and the iipper- 

 most wing-polyp is somewhat rudimentary. Quite similar features are found in the other specimens; 

 in some, the ventral polyp of the two-polyp wings is quite small, but a bud, and among the two- 

 polyp wings there may be a wing with only one polyp, but it, like the wings with two polyps, 

 is provided with two lateral zooids. At the apex of the stem one or two rudimentary polyps are found, 

 and above the uppermost there is generally a lateral zooid, which, together with the broken end of 

 the calcareous axis shows that the original apex must be wanting. 



Other specimens again, up to a length of 4o mm , show two-polyp wings quite to the apex with 

 two lateral zooids, and likewise in the rudiment-region, which in the larger specimens is long, and 

 reaches to the numerous stalk-zooids ; a specimen of a length of 40'" numbers on either side n 12 

 wings with developed polyps. 



A somewhat more advanced stage is represented by the specimen 45""" in length (from Bohuslan) 

 figured on PI. II, fig. 27. At the apex a naked end of the calcareous axis projects; uppermost are two 

 single, rudimentary polyps ; then follow two-polyp wings, of which the two uppermost are rudimentary, 

 whilst the others (5/6) are provided with two fully developed polyps; next follow wings with three 

 polyps the most ventral of which is the smallest, and the farther we get down, the smaller it becomes; 

 but so long as the polyp-buds of the smallest wings in the rudiment-region may be discerned, the 

 number continues to be three. Of lateral zooids three are found under a 1 1 the wings. The stalk- 

 zooids are mimerous (upwards of 16), and close-set, especially below. Similar features are seen in 

 several specimens, partly Swedish, partly from the Vestman Islands. Finally, we have a great number 

 of young colonies, 6o mm in length and others (Swedish, Icelandic specimens, one Danish fragment) in 

 which the developed wings and the rudiments, whose polyp-buds are distinct, contain three individuals; 

 at the apex, however, (if not damaged) a few two-polyp wings are regularly found, but with rudimentary 

 polyps. The number of lateral zooids is three or four; in the latter case one of them may be placed 

 in a second row. In the rudimentary wings where the most ventral polyp-bud is very small, the 

 corresponding lateral zooid is also indistinct. The rudiment zone is long and joins onto the line of 

 stalk-zooids; in a specimen (from Kristineberg) 6o ram long, with 15/15 fully developed wings I have 

 counted ca. 36 stalk-zooids in a closely packed series. 



The larger colonies with four, five, and more polyps in the wings, which are generally found 

 in collections, although most frequently in fragments, connect directly with such stages as the last- 

 mentioned. 



The young stages described accordingly show: i) that the number of polyps in the wings is 

 gradually increased by one individual in the rudiment-zone, 2) that by the longitudinal growth of this 

 zone the new wings advance farther upward, as also the older wings above with fewer polyps, 3) that 

 contemporaneously with this growth, transformations take place at the apex, as the oldest wings, and 

 upon the whole the older parts of the colony, here atrophy and finally drop off; for, the fact that 

 all the well-preserved young stages, even the smallest, are provided at the apex with indisputable 

 rudimentary polyps, without tentacles and often much diminished, seems to me to be most naturally 

 interpreted as an atrophy. A new light is thrown on the condition found also in the older colonies 

 by this interpretation: it is evident that the same processes continue, as long as the Virgularia is living. 



