Experimental data corresponding to the exoclastic fragmentsfrom spherical glass specimens fractured in gelatin in the course ofthe work on paper II are used to test the theoretical predictionsof paper IV; the fragments in question are those engendered bysurficial flaws of the specimen. The data on the fragments consistof measurements of the spherical surface corresponding to theoriginal surface of the specimen, and of the core surface. The lattersurface appears on many of the exoclastic fragments; it arisesthrough failure under compression of the central core of thespecimen to produce the endoclastic fragments. The over-allprobability of fracture determined only by the original sphericalsurface on the fragments shows the presence in the correspondingdistribution of three component distributions, corresponding ingeneral to the larger fragments, those of intermediate size, andthe smaller fragments. The first distribution is identified as arisingfrom fragmentation through activation of surficial flaws in thecore surface, and agreement of the data with theory is good.By eliminating fragments from the over-all distribution for whichthe data are correlated with theory in this manner, the seconddistribution in question is identified as corresponding to fragmentationthrough activation of surficial flaws in the original sphericalsurface of the specimen. Excellent agreement of the data with thetheory of IV is obtained. The remaining distribution probablyarises from particles formed by activation of internal flaws (althoughoriginal surface of the specimen appears on the fragments).The results yield numerical values of surficial flaw densities.