Sometimes you need to refer to a book or an article that you have read about in someone else's work, but not seen for yourself. If this happens then you should, if at all possible, try to locate the original material and cite it in the normal way.
If you are unable to find the original material then you may cite the work in your own text by referring to both the original and the citing authors.
For example, you could write in your assignment:
When human support is less available, attachment to pets is associated with better health (Garrity et al. (1989) cited in Tucker et al. (1995)).
And in your reference list you would put:
Garrity, T., Stallones, L., Marx, M. and Johnson, T. (1989) Pet ownership and attachment as supportive factors in the health of the elderly. Antrozos. 3, 35-44. Cited in Tucker, J., Freidman, H., Tsai, C. and Martin, L. (1995) Playing with pets and longevity among older people. Psychology and aging. 10 (1), 3-7.
Note that in the reference list the original work is written first, followed by the words Cited in, then the work where you found the information. The format of the two references follows the usual rules for the Harvard system, except that the title or journal title of the original work (Antrozos in our example) is not italicised; the title or journal title of the work you have read (the primary source) is italicised.
When referencing work cited by other authors, the original work is sometimes known as the secondary source and the work that cites it (i.e. the work that you are actually looking at) is the primary source. These terms should not be confused with the terms primary data and secondary data which have quite different meanings.