The branch also craved objective evaluation from outside its staff to gauge how well the exhibits it produced served their purposes. Behavioral scientists had developed two methods of conducting such research. One, involving close observation of a sufficient sampling of visitors, assumed that various measurable aspects of behavior reflected what went on in the minds of those observed. The other method used systematic questioning to assess quantitatively what a random selection of visitors took from the exhibits. The branch had some hope that park interpreters might engage in these studies and included a unit on exhibit evaluation in the annual Museum Methods Course. After they returned to their parks, however, few trainees attempted systematic studies of visitor response to exhibits.71 The branch saw one chance for a really professional study slip away but later established contact with an Office of Education project fostering exhibit evaluation research.

While scientific testing continued to elude its efforts, the branch did receive a flow of subjective comment that had cumulative impact. It solicited some of this from Carl E. Guthe, a highly respected practical museologist. Previously director of the New York State Museum, Guthe served the American Association of Museums as research associate from 1953 to 1959. In this capacity he crisscrossed the country with a house trailer studying particularly the problems of small museums. Ralph Lewis secured his appointment as a collaborator without compensation and invited him to visit and critique as many national park museums as he could in his travels.73 Guthe's reactions to the park museums he saw were consistently favorable, no doubt partly because the new visitor centers with their professionally designed and executed exhibits contrasted sharply with the majority of struggling small museums his studies involved.

Another source of outside evaluation tended to counterbalance this impression. Following completion in 1957 of the extensive developments at Colonial National Historical Park, the Service engaged a communications expert from academic circles to review the new installations. His pungently worded and aptly illustrated report identified numerous flaws ranging from the design of information desks to the architecture of auditoriums. In the museums he pointed out specific circulation difficulties, exhibit design concepts that failed, and specimen installations that did not fulfill their potential. His outspoken criticisms served to sharpen the eyes of Service personnel.

Most of the criticism directed at the exhibits in park museums came from within the Service. Carl Russell represented the viewpoint of material culture specialists and of collectors generally. He called on park museums to make greater use of historic objects in their exhibits and to label them more fully. Two Service colleagues echoed these recommendations when he made them in a paper before the Western Museums Conference in 1956. The plea for more specimens surfaced again in a discussion at the 1957 superintendents' conference. It was still being voiced strongly to the Western Historical Association at its 1963 meeting.

The Museum Branch had in fact placed considerable emphasis on specimens as evidence, illustration, and stimulus in its innovative development of narrative exhibits. It felt that injecting additional objects merely for their inherent interest would be a backward step. As for fuller labeling, Herbert Maier criticized park exhibits as having too much text.76 Between these contradictory views the branch strove to keep individual labels brief. It set 25 words as the desirable limit, which planners could not always achieve but at least approached. It trimmed drastically the label copy proposed by most park interpreters. Narrative exhibits as conceived by the branch did require fairly prominent title and key labels. These perhaps made the verbal content more obvious although not more lengthy.

Participants at the Chief Park Rangers' and Interpreters' Conference in March 1959 commented on the similarity in general appearance of park museum exhibits. This became the most consistently perceived fault of Mission 66 installations. As one planner phrased it, "The usual complaint, that the museums are all alike, we hear constantly."77 At the Visitor Services Conference in 1959, Maier and others raised the problem of visitors' difficulty in following the intended sequence of exhibits. Architect John B. Cabot mentioned this again in his 1960 paper before the Midwest Museums Conference. At the same time he decried Museum Branch preference for excluding or controlling daylight in exhibit rooms, for a single entrance/exit, and for counterclockwise circulation. He envisioned a grand collaboration of great designers from various fields to show museums the way out of such problems.