Abstract. Recent results on the yields of high-energy photo-protons from both bremsstrahlung and monochromatic photons incident mainly on carbon are compared with calculations using the Gabriel-Alsmiller model of high-energy photo-induced reactions. Good agreement is found for photons of energies up to 600 MeV.Recently, extensive sets of data on the cross-sections for nuclei from lithium to gold to produce high-energy photo-protons have become available [1,2]. In [1], data at an end-point energy of 400 MeV were given, while in [2], proton yields mainly from carbon irradiated with bremsstrah~ungofend-point energies between 125 and 1000MeV were listed, as were some data obtained by a "'photon difference" method. Some of the results of [1] have been analyzed in [3], in terms of the "quasi-deuteron" model as formulated by Matthews [4] and of the intra-nuclear cascade model (PICA) of Gabriel and Alsmiller [5]. In this paper, analysis of the data in [1] and [2] is extended, using the latter model. Three specific matters are examined:I. The earlier analysis indicated that, to obtain agreement with experiment, it was necessary to renormalize that part of the flux in the Gabriel-Alsmiller model which originates via the quasi-deuteron mechanism. The values of the quasi-deuteron constant, L, given by fits to the experimental data were _~ 6 for C, ~-8 for Cu and ___13 for Au. Although there thus appears to be a tendency for the value of L to increase with increasing atomic number, this conclusion depends largely on the value for gold. This is, however, the least well-determined of the values, partly due to the fact that the statistical accuracy of the calculations was not good and partly to the fact that the results for gold are not particularly sensitive to the exact value of quasi-deuteron constant used. (Few of the particles photo-produced in gold are protons.) Additional calculations have now been performed for Be and Ag, for comparison with the data of [I]. The calculated values have, as before, been fitted to experiment by treating the value of L as if it were a free parameter. The results are shown in Fig. 1. The values indicated by the comparison are L = 5 +1 for Be and L=10_+2 for Ag, (As in the earlier analysis, these values are determined by the value of L giving the best visual fit to the experimental data. The errors quoted in the values of L do not take into account any errors in the input data, e.g. in the values of the cross-sections for deuteron photo-disintegration. These are usually smaller than the statistical errors in P1CA calculations.) With the present results, PICA analyses have now been performed for six of the elements for which data were reported in [1] (A1 has been studied by Gabriel who finds L=7.2 [6]). All the analyses confirm the trend reported in [1], that, within the framework of the Gabriel-Alsmiller model, the value of quasi-