3i6 The Irish Naturalist* December, 



the hybrid, b, both grew rarer and rarer, until the}^ almost 

 ceased at the bathing-place. Thence for nearly a quarter of a 

 mile farther south-westward along the steep cliffs capped with 

 drift, both S.JacobcEa and the hybrid, a {S. albescens^ continued 

 to appear in association, S. Cineraria being quite absent from 

 the latter half of this distance. The inland extension of 

 S. albescens was small. Several plants .appeared in Sorrento 

 Park above the Cliffs, others appeared by the roadwaj^ above 

 the railwa}^ at Vico, others again near Cooliemore Harbour, 

 about a furlong north of the Cliffs — some even on the summit 

 of the hill above Kh3^ber Pass — S- Jacobcsa in all of these sta- 

 tions occurring close b}'. But the headquarters of .S. albescens 

 were along the railway-banks and sea-banks at and near Vico 

 bathing-place, and along the cliffs and banks b}^ the sea to the 

 south-westward. 



On the assumption of a hybrid origin for the intermediates, 

 it seems at first rather hard to account for their peculiar 

 absence from Sorrento Cliffs, where one of the parents grows 

 in such abundance, and at some points within a stone's-throw 

 of the other parent. The pollen of both of the assumed parent 

 species is equall}^ adapted for wind or insect carriage. Cross- 

 fertilization must hav^e been effected by either agenc}' ; and it 

 seems just as inadmissible to assume the winds to have blown 

 persistentl}' from S> Cineraria towards S- Jacobcea as to assume 

 the honey-seeking insects (bees, most probabl}') to have invari- 

 ably travelled in the same direction. Both the winds and the 

 bees must have frequently travelled in the reverse direction, 

 carrying the pollen of S.Jacobcea to the stigmas of 5. Cineraria, 

 perhaps as often as they carried the pollen of the latter species 

 to the stigmas of the former. And this being so, does not 

 the absence of intermediates, it may be argued, from the cliffs 

 where S. Cineraria reaches its maximum, show that the sug- 

 gested formation of hybrids does not reall}' take place. If we 

 assume, however, what has long since been proved for other 

 species capable of producing hybrids, that there is a want of 

 reciprocity of cross-fertilization between S. Cineraria and 5. 

 Jacobcea^ the difficulty disappears. The pollen of 6". Cineraria 

 may be able to fertilize the ovules of S. Jacobcea, while the 

 pollen of the latter is inert as regards the ovules of the former. 

 There may be, in short, a perfectly free interchange of pollen 



