1 Introduction

The low-excitation transitions of planetary nebulae have long constituted a thorny problem for nebular spectral analysis and model simulations. Thus, whilst the intensities of intermediate to high excitation lines can often be represented in terms of quite simple nebular models (e.g. Aller 1982), lower excitation line simulations are frequently grossly in error (cf. Köppen 1983; Middlemass 1990); a disparity which has only partially been alleviated through recent use of more realistic (non-homogeneous, non-spherically symmetric) source structures (Hyung et al. 1994, 1995; Hyung & Aller 1996). Whilst the reasons for these differences are far from clear, it appears likely that "excess" components of emission are associated (in part) with micro-filamentary structures and clumps (Hyung et al. 1994, 1995; Köppen 1979; Boeshaar 1974; Hyung & Aller 1995).

A further complicating factor relates to the possible contribution of shocks; a process which is believed to be important for explaining certain small (e.g. Balick et al. 1993, 1994; although see later) and larger scale nebular structures (Cuesta et al. 1993, 1995; Balick et al. 1983; Icke & Preston 1989; Icke et al. 1992), and would likely result in enhanced low-excitation transitions (Shull & McKee 1979; Raymond et al. 1989; Hartigan et al. 1987). Distinctive shock emission signatures have already been noted in for instance CRL 618, M2-56 (Goodrich 1991), KjPn 8 (Lopez et al. 1995) and a variety of other sources (Bohigas 1994; Balick et al. 1993, 1994; Meaburn & Walsh 1980a, 1980b; Rowlands et al. 1994; Trammell & Goodrich 1996), although the general prevalence of shocks (and their importance in explaining lower excitation line strengths) remains far from clear.

In the following, we shall analyse a broad range of published optical spectra with the aim of evaluating trends in (de-reddened) line emission for some 538 nebulae. These are compared with the variations expected for radiative and shock excitation models in an attempt to evaluate the importance of collisional excitation; the extent to which low excitation line anomalies are representative of planetary nebulae (PN) in general; and finally, to determine the evolution in relative selected line ratios as a function of nebular radius.