[…] a zero segment in a given environment can only be a member of one class. Defining a zero segment may be useful in a case such as the following: Suppose that the sequences 'AXaYa', 'AXbYb', and 'AXc' occur (where 'Xa' and 'Xb' are either the same or else different morphemes of class 'X'), and that 'Xa' and 'Xb' have no descriptively relevant difference as against 'Xc' except for the relation to 'Y' stated here. Then we recognize a zero segment after 'AXc' as a member ('Yz') of 'Y', and thus obtain the element sequence 'AXcYz'. We can now say that each occurrence of the environment 'AX'—contains some member of other of 'Y'. Techniques of this type are especially useful when we wish to set up 'AXY' as a construction […] and do not wish to exclude therefrom the 'AXc' sequence. Examples of such zero segments in phonology were the phonemic junctures […], which set up a new phonemic element indicating a zero segment in a unique environment of other segments. The juncture elements do not, of course, represent zero; they represent particular features of neighboring segments. But the position they occupy is that of a segment of zero length in the utterance.
- Harris (1951), a pag.335

[…] both zero segments (including junctures) and void elements are representations of sequences of segments, as are also phonemic and morphemic components and, if we will, the resultant class of chapter 16. The only status that the symbol {'-en'} has in the representation 'NVbV-en' for 'I have cut' is what can be extracted from the difference in the 'NVbV-en' representation of 'I have cut' and the 'NV' representation of 'I cut'.
- Harris (1951), a pag.370, n.13