2-methylpropan-2-ol-4 25N-ammonia (4:1, v/v). Many solvent systems for these compounds have been proposed (see, for example, Kubota, 1961); the information is given because the system seems to be a very versatile one. It was used (Vining & Taber, 1957) for oc-amino acids and c-hydroxy acids, and later (Russell, 1960) for N-methyl-amino acids. It is a stable one-phase system, and gives compact spots with DNP-derivatives, the R. values lying between 0.1 and 0-8 approximately. The DNP-derivatives of leucine and isoleucine are particularly well separated. SUMMARY 1. Solutions of DNP-amino acids and-peptides in dilute aqueous sodium hydrogen carbonate were examined photochemically. 2. The a-DNP-amino acids were all rapidly decomposed by light with the exception of DNPtryptophan; $., 8-and c-DNP-amino acids were stable, and DNP-peptides were either stable or decomposed much more slowly than a-DNP-amino acids. 3. Formation of 4-nitro-2-nitrosoaniline on photolysis appeared to be a property exclusively of (lerivatives possessing the structure DNP *NH* CRR' CO2H, but not all compounds possessing this structure yield 4-nitro-2-nitrosoaniline. 4. The solvent system 2-methylpropan-2-ol-4 25N-ammonia (4:1, v/v) was useful for the paper chromatography of DNP-amino acids.

Measurements of CSF-albumin, and S/CSF-albumin in two groups of patients with small and large infarcts failed to show BBB damage to albumin in about one half of the patients. N o significant difference was found between the two groups of patients in regard to the above-mentioned parameters. The localization of the infarcts (close to the CSF or deep in the brain) did not influence the S/CSF-albumin. No correlation was found between the S/CSF-albumin and the time interval between the onset of the stroke and the lumbar puncture within 96 h. Some suggestions are put forward to explain the discrepancies between the CTscan and S/CSF-albumin in the assessment of BBB damage. K e y words: Blood-brain barriercerebral infarctioncerebrospinal fluid -CSF-albumin -CT-scan -S/CSF-albumin.

Cerebrospinal fluids from patients with neurological diseases have been studied. It appeared that the detectability of immunoglobulin G subclasses 1, 2, 3, 4, and the occurrence of oligoclonal bands in the IgG region increased with increasing concentration of IgG in the spinal fluid. No specific pattern emerged in the different disease groups. The reference intervals were determined for: Sp-Albumin(66,OOO): 0.08-4.4 pmol/l; Sp-IgG(160,OOO): 0.06-0.26 pmoV1; S-Albumin : 120413, Sp-AlbuminSp-IgG X S-Albumin within the reference interval, while S-IgG Sp-Albumin was increased. Sp: spinal fluid S-AlbuminSp-IgG X S-Albumin Sp-Albumin S-IgG X Sp-Albumin In these ratios the variability of the concentration of albumin and IgG in serum is taken into account and corrected for.Reference intervals and the values obtained in a variety of neurological diseases for these parameters are presented. and MATERIAL Reference interval samplesCerebrospinal fluid specimens with Sp-Protein (ref. int. 0.15-0.50 g/l), Sp-Erythrocytes and Sp-Leucocytes (ref. int. 0-3 X 106/1) within the reference interval were used to determine the reference intervals for Sp-Albumin(66,000) Sp-Immunoglobulin G S-Albumin Sp-IgG X S-Albumin (160,000), Sp-Albumin and S-IgG Sp-Albumin according to Delpech & Lichtblau (1972) Spinal fluid controlsA pool of 60 reference interval samples of spinal fluid was prepared. The Sp-IgG was 0.15 pmol/l and the Sp-Albumin was 3 pmolll. Part of the pool was concentrated 10 times by dialysis against Carbowax 20 M, and nonconcentrated as well as 10 times concentrated pool was used as control. The pools were stored at -20°C in portions of 1 ml.

No abstract