According to Ingram, the words marked (2) are to be regarded as light endings ‘only when used as auxiliaries’; those marked (3), ‘when not directly interrogative’; those marked (4), ‘when followed immediately by as.’ Such belongs to this class, ‘when followed by a substantive with an indefinite article, as Such a man.’ There are hardly any weak or light endings in the first and second periods of Shakespeare’s work. In the third they occur now and then and become more frequent in the last period. So we have e.g. in Antony and Cleopatra (1600) 3·53 per cent.; in The Tempest (1610) 4·59 per cent.; in Winter’s Tale (1611) 5·48 per cent.
In the application of this test we must chiefly keep in mind that these two groups of words are only to be considered as ‘weak’ and ‘light’ endings when they form the last arsis of the line, as is the case in the lines quoted from Henry VIII; but they are to be looked upon as part of a disyllabic or feminine ending if they form a supernumerary thesis following upon the last arsis:
Upón this groúnd; | and móre it woúld contént me.
Wint. II. i. 159
§ 168. Intimately connected with the quality of the line-endings is the proportion of unstopt or ‘run-on’ and ‘end-stopt’ lines, or the frequent or rare use the poet makes of enjambement. Like the feminine, weak, and light endings, this metrical peculiarity also occurs much more rarely in Shakespeare’s earlier than in his later plays. According to Furnivall’s statistics, e.g. in Love’s Labour’s Lost one run-on line occurs in 18·14 lines; in The Tempest, on the other hand, we have one run-on line in 3·02 lines; in Winter’s Tale the proportion rises to one in 2·12.
As in the later plays run-on lines are often the result of the use of weak and light endings, we may perhaps assume with Hertzberg that at times the poet deliberately intended to give a greater regularity to the verse, if only by introducing the more customary masculine endings. From this point of view, then, both the weak and light endings and the run-on lines would have much less importance as metrical and chronological tests than they otherwise might have had.
§ 169. But there is another peculiarity of Shakespeare’s rhythms noticed by Hertzberg which is of greater value as a metrical test; viz. the use of the full syllabic forms of the suffixes -est, and especially of -es or -eth in the second and third pers. sing., as well as that of -ed of the preterite and of the past participle. These tests are all the more trustworthy because they do not so much arise from a conscious choice on the part of the poet as from the historical development of the language. This is indicated by the fact that the slurring of these endings prevails more and more in the later plays.
According to Hertzberg’s statistics the proportion of fully sounded and slurred e is as follows:
| 1 H. VI. | T. Andr. | 1 H. IV. | H. VIII. | |
| 3 Pers. Sing. | 15·58% | 6·4% | 2·25% | 0% |
| Pret. and P.P. | 20·9% | 21·72% | 15·41% | 4·2% |
It thus appears that in this respect also there is a decided progress from a more archaic and rigorous to a more modern usage.