Composite materials often fail by delamination. As composite materials with tougher matrices are developed to give better delamination resistance, their delamination behavior needs to be fully characterized. In this paper the onset and growth of delamination in AS4/PEEK, a tough thermoplastic matrix composite, was characterized for Mode I and Mode II loadings, using the double-cantilever beam (DCB) and the end-notched flexure (ENF) test specimens, respectively. Delamination growth per fatigue cycle, da/dN, was related to strain energy release rate, G, by means of a power law. However, the exponents of these power laws were too large for them to be adequately used as a life prediction tool. A small error in the estimated applied loads could lead to large errors, at least one order of magnitude, in the delamination growth rates. Hence, strain energy release rate thresholds, Gth, below which no delamination would occur were also measured. Mode I and II threshold G values for no delamination growth were found by monitoring the number of cycles to delamination onset in the DCB and ENF specimens. The maximum applied G for which no delamination growth had occurred until at least 106 cycles was considered the threshold strain energy release rate. The Gth values for both Mode I and Mode II were much less than their corresponding fracture toughnesses. Results show that specimens that had been statically precracked in shear have similar Gth values for Mode I and Mode II for R-ratios of 0.1 and 0.5. An expression was developed which relates Gth and Gc to cyclic delamination growth rate. Comments are given on how testing effects (e.g., facial interference and damage ahead of the delamination front) may invalidate the experimental determination of the constants in the expression.