SYNOPSIS (4) 1.3.2. [Vārttika] If the term dhātu is assigned to an item by virtue merely of its being recited in the dhātupāṭha, another kind of provision has to be made that such and such a particular stretch of sound constitutes a separate dhātu (vt. 2: parimāṇagrahaṇaṁ ca). In his Bhāsya on this Vārttika, Patañjali states: "But for such a provision, one might consider the sequence bhvedha a single unit, instead of considering bhū and edhA as two distinct units." This statement has been taken as showing that the original dhātupāṭha contained no meaning entries (arthapāṭha). A few points of importance must be set out:, 1) Pārimāṇagrahaṇa ('mention of the measure') is required as far as bhū and edhA are assumed to have no meanings whatsoever. There do not really arise problems of this sort in determinig what are the units in the dhātupāṭha, since in Pāṇini's grammar the units in the text are postulated as individual, meaningful ones, like the members of a cādi-group. What is meant by Patañjali here is that one cannot make the segmentation of a given sound-sequence into its constituents without taking into account the meanings to be expressed by them; meanings are essential for dividing a given chain of sounds into the meaningful units. 2) Therefore, from the context of the Bhāṣya one has to learn that the question raised is how to mark off the sequence bhvedha without bringing in the meanings. Only on the assumption that the root entries are given as distinct, meaningful units in the dhātupāṭha, one may talk about Pāṇini's way of reciting them. 3) Meaning entries, as they from a part of the dhātupāṭha, discriminate (parcchedaka) given root entries, not in such a way that a meaning entry intervenes between two root entries, but in the way that it serves to know a given root entry to be a meaningful unit, that is, to provide the ground for the segmentation. As far as I know, no scholars interested in the dhātupāṭha noticed this point. 4) Considering that Pāṇini refers to root entries by saying bhūvādayaḥ in the present rule, it seems to be reasonable that Patañjali speaks of bhvedha, since by using ādi a set of items to be called dhātu is referred to. 1. 4. 1. New formulation of the definition rule for the term dhātu is proposed: kriyāvacano dhātuḥ ('a speech unit that signifies an action is called dhātu'), as opposed to the definition based on the dhātupāṭha (pāṭhena dhātusantjñā). (To be continued.)