Abdokov (1983, 131-132) compares the EC forms with Kab. -ṣ́ǝ- in wǝ-ṣ́ǝ-n 'to open, open wide', xʷa-ṣ́ǝ-n 'to close, shut'. The root is probably connected with Ad. -č̣́ǝ- in za-ʡʷǝ-č̣́ǝ-n, q:ǝ-ʡʷǝ-č̣́ǝ-n 'to open (intr.)' and thus allows to reconstruct PAK *č̣́ǝ-. The phoneme *č̣́ theoretically can go back to PWC lateral *ƛ̣́ʷ, thus allowing for a remote chance of relationship with the EC root; however, it seems much more probable that together with Ub. č̣́a- (in č̣́a-tʷ- 'to close', č̣́a-ṭʷ- 'to open') it goes back to a quite separate WC protoform *č̣́V 'to open / close'.