succeed in laying a coat of European varnish over the specifically Prussian bureaucrat. How these observations acted in practice is clearly shown when we go through the list of our diplomatists of those days: one is astonished to find so few native Prussians among them. The fact of being the son of a foreign ambassador accredited to Berlin was of itself ground for preference. The diplomatists who had grown up in small courts and had been taken into the Prussian service had not infrequently the advantage over natives of greater assurance in Court circles and a greater absence of shyness. An especial example of this tendency was Herr von Schleinitz. In the list we find also members of noble houses in whom descent supplied the place of talent. I scarcely remember from the period when I was appointed to Frankfort anyone of Prussian descent being appointed chief of an important mission, except myself, Baron Carl von Werther, Canitz, and Count Max Hatzfeldt (who had a French wife). Foreign names were at a premium: Brassier, Perponcher, Savigny, Oriola. It was presumed that they had greater fluency in French, and they were more out of the common. Another feature was the disinclination to accept personal responsibility when not covered by unmistakable instructions, just as was the case in the military service in 1806 in the old school of the Frederickian period. Even in those days we were breeding stuff for officers, even as high as the rank of regimental commander, to a pitch of perfection attained by no other state; but beyond that rank the native Prussian blood was no longer fertile in talents, as in the time of Frederick the Great. Our most successful commanders, Blücher, Gneisenau, Moltke, Goeben, were not original Prussian products, any more than Stein, Hardenberg, Motz, and Grolmann in the Civil Service. It is as though our statesmen, like the trees in nurseries, needed transplanting in order that their roots might find full development.
Ancillon advised me first of all to pass my examination as Regierungs-Assessor, and then, by the circuitous route
