Appropriately for a poet who is ‘subtly speaking’ (λεπτολόγος), the epithet applied to him by Ptolemy III Euergetes (Suppl. Hell. 712.4), Aratus does not cease offering unexpected material to explore. This statement holds true also for the famous passage containing the acrostic ΛΕΠΤΗ (lines 783–7):

In modern times, its hidden layers were detected gradually. For centuries, the students of the Phaenomena were aware only of what was visible while reading horizontally, that is, of the two instances of ΛΕΠΤΗ inserted, respectively, in lines 783 and 784. Significant progress was made late, in 1960, by Jean-Marie Jacques,Footnote ² who was the first to notice that the initial ΛΕΠΤΗ appears both horizontally and vertically (as an acrostic). Thus, since then the number of the identified occurrences of the adjective in question within these five lines has increased to three, and this was the established state of the art for the next half a century.

The intentionality of Aratus' literary game was fully proven by these purposely accumulated adjectives. There may be an oblique comment on the situation after Jacques's publication until quite recently in Aratus' ipsissima verba, i.e. the lines quoted above (καλὸν δ᾽ ἐπὶ σήματι σῆμα | σκέπτεσθαι, κτλ.). If we take them, tentatively, as a general instruction which is meant to direct retrospectivelyFootnote ³ the reader's attention not only to the secrets of the sky, but also – on the metapoetic level – to the text itself, the three instances of ΛΕΠΤΗ can indeed, in respect of their number, provide the feeling of satisfactory completeness.

On close inspection, however, it appears that the trinum identified by Jacques does not exhaust all the hidden resources of the text. As has been demonstrated recently by Mathias Hanses,Footnote ⁴ in the passage under discussion one can discern the fourth ΛΕΠΤΗ, consisting of the first letters of the words in consecutive lines, disposed in a deliberate diagonal arrangement:

Mathias Hanses' unconventional approach inspired me to seek further attestations of Aratus' cryptic art in this passage which had seemed thoroughly explored. I think that further progress can be made when we ask the question about the possible ‘starting points’ of acrostics (or other kindred forms). The acrostic detected by Jacques (an example of the so-called ‘gamma-acrostic’)Footnote ⁵ is modelled on the traditional pattern known from funerary inscriptions,Footnote ⁶ whereas the one proposed by Hanses anticipates the combinatory verses of Optatian Porfyry.Footnote ⁷ What the two patterns have in common is that both start from the first letter of the first (initial) ΛΕΠΤΗ. But what about the second ΛΕΠΤΗ? Does it serve only to verify the intentionality of the ‘frontal’ acrostic or is it charged, additionally, with yet another function? I am convinced that the latter possibility is at least worth considering.

Let us give a closer look at line 783. In its second half, there occurs the phrase περὶ τρίτον ἦμαρ, placed just after the caesura of the type usually called κατὰ τρίτον τροχαῖον. The initial letters of that phrase spell ΠΤΗ, which does not seem to be accidental, bearing in mind the immediate context and the fact that Aratus as a rule avoids acronyms in his poem (hence any exception becomes meaningful).Footnote ⁸ The missing part is to be found in the following line 784, this time around the caesura penthemimeres. To obtain the whole ΛΕΠΤΗ one has to read boustrophedon Footnote ⁹ the first letters of the appropriate words (those underlined in the quotation below), starting from the lower line (right-to-left) and continuing in the upper line (left-to-right).Footnote ¹⁰ This kind of layout of the inscribed text resembles the so-called boustrophedon ab imo which was used in some archaic inscriptions (the parallel applies exclusively to the direction of reading the first letters). Here are the lines under discussion; the arrows indicate the required direction of reading:

The ‘internal’ signposting technique employed by Aratus consists in suggesting the direction of reading by the use of the built-in expression περὶ τρίτον ἦμαρ, which in itself – when interpreted literally – may imply the meaning ‘around the third day’, that is, in a circular way (recalling the twists of boustrophedon inscriptions) within the description of the moon on the third day. The intentionality of the pattern is guaranteed by positioning ΛΕΠΤΗ at the very beginning of the sequence, just as it happens in the case of the two acrostics so far discerned. The appropriateness of this acronym to its contextFootnote ¹¹ seems obvious on account of the double repetition of the pivotal adjective in this particular couplet.

What is more, there are further probable signposts pointing metapoetically to this acronym. As it is easy to see, its former part appears in the first half of line 784, the latter – in the second half of line 783, which is in tune with the stress put on observing the moon not only at full, but also at the two halves (799–809), which finally leads to emphasizing these aspects by the acrostic ΠΑΣΑ (803–6)Footnote ¹² followed by ΜΕ-ΣΗ (807–8).Footnote ¹³ The signs occurring on the third (and fourth) days are to be differentiated according to the rule of dichotomization (for example, up to mid-month and after mid-month), which in Aratus' poetic code may be understood as a reference to the halves of the hexameter. And, as we have seen, ΛΕ is contained in the first half of line 784 (up to its mid-point), and ΠΤΗ in the second half of line 783 (just after the mid-point).

In conclusion, Aratus inserts in the passage in question, in addition to what has already been detected by the scholars, a sequence of words arranged to form an acronym spelling ΛΕΠΤΗ when read boustrophedon. This device, like the other ones, does not come as a complete surprise. A little earlier, at 778–9, the reader is asked, metapoetically, to observe (σκέπτεο) on either side (ἑκάτερθε)Footnote ¹⁴ what the poet inscribes (ἐπιγράφει);Footnote ¹⁵ the shape of the inscription may be different at different times (ἄλλοτε ... ἄλλῃ ... αἴγλῃ). Such a loose poetic formula encapsulates all the variants of Aratus' ΛΕΠΤΗ tricks mentioned above.Footnote ¹⁶