THE SUBDIVISION OF lEELAND xxxix 



III. THE SUBDIVISION OF IRELAND. 



The plan adopted in this work, according to which Ireland is 

 partitioned into forty county-divisions, is, except as regards the 

 County of Donegal, that which was proposed by me in the "Journal 

 of Botany" and the "Irish Naturalist," for February, 1896, in a 

 paper entitled ' On the botanical Subdivision of Ireland.' This 

 scheme is a modification of one put forward by the late Prof. 

 Babington, of Cambridge, in a paper, ' Hints towards a Cybele 

 Hibernica,' which was read before the Dublin University Zoological 

 and Botanical Association, on February 18, 1859, and published in 

 the " Natural History Eeview" of that year (vol. vi., part 2, pp. 533- 

 537). Babington's scheme was an adaptation to Ireland of H. 

 C. Watson's subdivision of Great Britain, as first published in 

 " Cybele Britannica," vol. iii., 1852, and used in detail in his 

 " Topographical Botany," 1873-74. 



The botanical divisions correspond, in the main, with the Irish 

 counties. It is clearly desirable that the unit of area employed in 

 Great Britain and in Ireland respectively should correspond, for 

 purposes of comparison and of general census ; the size of the 

 majority of the Irish counties is favourable to this object. The 

 six largest counties are sub-divided, forming fourteen divisions of a 

 little over average size. None of the counties are so small as to 

 render it necessary to group them, as was done by Watson in one 

 or two cases in Great Britain. 



In working out the boundaries of the divisions, the following 

 considerations appeared to be important : — 



Natural boundaries. — Where clearly-defined natural boundaries, 

 botanical, geological, or physical, exist, it is manifestly advan- 

 tageous that they should be followed ; but it is not always 

 possible to follow them, on account of other considerations. The 

 convenience of county-divisions is so great, that except in the 

 subdividing of a large county, it does not appear desirable to forsake 

 county boundaries. 



Nature of boundaries. — Where a new boundary-line is required, 



