T. E. Page’s edition (with commentary) of Virgil’s works (first edition 1898) is enormously useful. (The volume on the Bucolics and Georgics is available here). Here is an illustration of Page’s singular philological prudence.
He considers Georgics 2.140–144 (Virgil refers to Jason’s sowing of a dragon’s teeth):
haec loca non tauri spirantes naribus ignem
inuertere satis immanis dentibus hydri,
nec galeis densisque uirum seges horruit hastis;
sed grauidae fruges et Bacchi Massicus umor
impleuere; tenent oleae armentaque laeta.
Other commentators, he notes, have “raised a difficulty” over the phrase satis … dentibus, which can be understood as an ablative absolute. The difficulty is that “as ploughing precedes sowing … you would not plough the land ‘when the teeth had been sown.'” (The perfect tense of the participle satis would typically indicate time prior to that of the main verb invertere.) To escape the (apparent) difficulty commentators have resorted to various explanations: It is a hysteron proteron; or a dative approximating to either serendis dentibus (Wagner) or propter sationem dentium (Madvig)—”for the sowing of (a dragon’s) teeth.”
Not page. Here is his explanation:
The strong simplicity of the passage, however, precludes these artificial explanations, and the obvious rendering is the right one—’these lands no fire-breathing bulls ever ploughed when dragon’s teeth were sown’ or ‘at the sowing of a dragon’s teeth.’ In referring to a remote mythological event the minute question as to which of two acts preceded the other is not in Virgil’s mind, but he mentions the two as jointly constituting one event. … Moreover, the past sense of the past participle is not always prominent, and the rendering ‘when dragon’s teeth were being (not ‘had been’) sown’ is perfectly justifiable.