side. 'Singing hymns' suggested Fiommar; 'and excommunicating the sharks,' added her brother."
But it would be unfair to leave the impression that the average level of Mrs Mitchison's writing is as vulgar and silly as this. On occasions she can do far better, as for instance when she so happily puts us in touch with the rude hilarity of the Gallic rabble who have treacherously taken Titus prisoner while he is buying grain.
"One of them rode up with a sack of corn across his horse’s back behind, a bag of gold in front; he shouted, 'I've been marketing with the Romans!'"
That is excellent. As we read it we know it is just what might have happened. Indeed this art of plausibly reconstructing the mood and atmosphere of the dead past is a most subtle one. Strangely enough any attempt at exact realism almost inevitably destroys the sense of reality, the best effects being won, it would seem, by vagueness and omission. The simple allusion to Fiommar "putting on her clothes" reveals an injudicious use of words; still
more, perhaps, is this so when we are told that the "blood was sticky on her dress." In both cases more uncertain words should surely have been selected, words that would have left the imagination free from too familiar associations.
However, in those backgrounds of her story which reflect certain aspects of nature, where infallibility is as a rule assumed, her work is often marred by inaccuracy of observation. Lerrys, the Gaul, is led to a Druidical Grove in Britain to take a false oath, but one's conviction as to the reality of the scene is in no way increased by being told that the "foxgloves rocked in the breeze" when any one who is familiar with the woods of England in the late summer knows that these particular flowers have long passed their time of blooming by the latter end of August. In the same way our sense of the reality of Meromic's journey from France to Italy is decidedly diminished by the information that "Overhead the new, bright stars crowded the sky." In so short a migration southward it would require an extremely punctilious astronomical observer to note any difference in the constellations at all.
