The nutritional status of a cancer patient may be affected by the tumor, the chemotherapy and/or radiation therapy directed against the tumor, and by complications associated with that therapy. Chemotherapy-radiotherapy is not confined exclusively to malignant cell populations; thus, normal tissues may also be affected by the therapy and may contribute to specific nutritional problems. Impaired nutrition due to anorexia, mucositis, nausea, vomiting, and diarrhea may be dependent upon the specific chemotherapeutic agent, dose, or schedule utilized. Similar side effects from radiation therapy depend upon the dose, fractionation, and volume irradiated. When combined modality treatment is given the nutritional consequences may be magnified. Prospective, randomized clinical trials are underway to investigate the efficacy of nutritional support during chemotherapy-radiotherapy on tolerance to treatment, complications from treatment, and response rates to treatment. Preliminary results demonstrate that the administration of total parenteral nutrition is successful in maintaining weight during radiation therapy and chemotherapy, but that weight loss occurs after discontinuation of nutritional support. Thus, long-term evaluation is mandatory to learn the impact of nutritional support on survival, disease-free survival, and complication rates, as well as on the possible prevention of morbidity associated with aggressive chemotherapy-radiation therapy.

No abstract

RITTEK, S. M. The vertical-horizontal illusion. An experimental study of meridianal disparities in the visual field.

No abstract

Sequences containing 10 numbers from 1-20 were presented in serial-order or random-order A previous study concerned with the involvement of long-term storage in encoding for short-term storage showed that when numbers were presented in serial-order (either lowest to highest or the reverse) serial-order recall (lowest to highest) was greater than when presentation was in random-order (Buschke, 1968;Buschke & Mesibov, 1967). Greater recall of serial-order presentation was attributed to a decrease in the range of alternatives in long-term storage which are relevant for identification and encoding when presentation is in serial-order, so that more rapid identification might provide additional time for encoding more information.The purpose of the present investigation was to evaluate that interpretation further by comparison of recall at increasing rates of presentation, as the interpresentation interval for identification and encoding decreases while stimulus duration remains constant. If greater serial-order recall of serial-order presentation (L-+H/L-+H) than of random-order presentation (Rand/Ls-H) were due to reordering or repeated sequential scanning of presentations in Rand/Le-H, recall should decrease more in Rand/Lr-H than in L-+ H/L-+H as rate of presentation increases. If the decrease in recall for Rand/L» H were similar to that for L-+ H/L-+H it would indicate that Rand/L» H does not involve reordering or sequential scanning. If recall for L~H/L~H is greater because identification is more rapid, recall for L-+H/L~H should not decrease as much as recall for Rand/L» H if identification and encoding are equivalent (so that identifying is encoding). However, if encoding followed identification and if mean identification time were constant, then recall for L-+ H/L-+H should decrease as much as recall for Rand/L» H, since the amount of additional encoding time for L~H/L-+H would be constant at all rates of presentation. METHOD SubjectsThe Ss were eight college students who were paid for their participation in two I-h test sessions. MaterialsThere were 20 sequences of 10 numbers chosen randomly from the set 1-20 for each of the four conditions, at each of the three rates of presentation. The sequences were presented either in random-order or in natural serial-order (from lowest to highest number), for serial-order recall (from lowest to highest) or same-order recall (in order of presentation). The four conditions resulting from two kinds of presentation and recall were: L~H/L~H, Rand/L~H, L-+ H/Same, and Rand/Same. All numbers were presented equally often, and in random-order presentations each number occurred about equally often in each presentation position. Procedure Each S was tested individually in an lAC ACT-1202 sound room, viewing sequences of numbers through a glass window. The numbers were shown by an lEE Bina-View which was programmed by a teletype tape reader.Each sequence was preceded by 3 sec of zeros presented at 4/sec and was followed by 20 sec of 4/sec blanks. At all rates of presentation eac...