 |
 |
 |
Ruskin provides his own definition of ‘roll’ at Works, 9.328 - ‘any mouldings which present to the eye somewhat the appearance of being cylindrical’. They are, he claims to be ‘considered as generally the best means of drawing an architectural line, the soft curve of their side obtaining some shadow at nearly all times of day...’
At Notebook M p.20 the reading is presumably ‘roll’, though the Ruskin Library Transcript T7A reads ‘rolt’. The second ‘l’ is awkwardly formed but ‘roll’, unlike ‘rolt’, makes sense here and seems preferable.
|
|
|
[Version 0.05: May 2008]