rates and approaches the amorphous state with very rapid quenches. These changes in structure cause changes in many of the magnetic, mechanical, optical, and thermal properties. As has been demonstrated in the Y-Ir system, it should be possible to prepare eutectics with regions of crystallinity large enough to keep the N(0) large and sharp, but small enough to ensure that the Debye temperatures are lowered dramatically. Such preparations could lead to much higher Te materials and provide a tremendous impetus to the widespread application of superconductivity.

Understanding Eutectics

The results of the present study may prove to be even more important outside the field of superconductivity. In metallurgy, eutectics have been known and studied for a long time without real progress in understanding when they occur and why they have such altered properties. The lowering of a melting point by mixing one substance with another has been known for a long time-but not understood. No theory or even hypothesis will predict the occurrence of eutectics. Now, the enhancement of T for the first time opens the way to a basic understanding of eutectics. We now realize that lattice softening goes hand in hand with the melting point minimum in the eutectics. We now have an important new piece in the puzzle of eutectic behavior.

The world has relied on eutectics for a long time, from getting rid of ice with salt to making soft solder, and from welding, to alkaline liquid coolants in reactors. Finally, some understanding is on the way; with it comes the possibility of progress toward higher Te, technologically more useful superconductors, and perhaps eventually a better understanding of many metallic alloys.

degree in chemistry in 1939, came to LASL in 1946, and received his Ph.D. in physical chemistry from the University of New Mexico in 1956. He has over 30 years of varied experience in research, development, and instrumentation in the fields of electrochemistry, radiochemistry, high-temperature phase studies, and cryogenics. His continuing investigations of superconducting and magnetic properties of various carbides and intermetallic compounds have resulted in discoveries of several new superconductors, a new itinerant electron ferromagnet, and the first itinerant antiferromagnet.

c

Gregory R. Stewart earned his bachelor of science degree cum laude from California Institute of Technology in 1971, and received his Ph.D. from Stanford University in 1975. He did postdoctoral work at Stanford and at the University of Kostanz, FRG. He joined LASL's CMB Division as a Staff Member in 1977, where he developed a new type of small-sample calorimeter and constructed a 7-T superconducting magnet. He is involved in characterizing new materials that range from high T, superconductors of technological interest to lower T, superconductors, where the interest is in understanding what makes superconductivity occur. James L. Smith is best known for his definitive work on three of the five known superconducting actinides. He also has studied high-pressure phase transitions of americium. As an offshoot of the work described in the article, he has made interesting discoveries about dilute magnetism; for example, he has discovered magnets that are 100 times more dilute than any found before. He earned his bachelor of science degree at Wayne State University in 1965 and his Ph.D. in physics from Brown University in 1974. He has been a Staff Member in LASL's CMB Division since 1973.

Bernd T. Matthias is Associate Director of the Institute for Pure and Applied Physical Sciences and Professor of Physics at the University of California at San Diego, and part-time member of the technical staff of Bell Telephone Laboratories, Inc. He has been a consultant for LASL since 1957 and was named a Fellow of the Los Alamos Scientific Laboratory in 1971. He is a world authority on superconducting materials, having been involved in the discovery of more of these than any other individual. He is also well known for his discoveries of ferroelectrics and ferromagnets, and for his work on the questions of the coexistence of magnetism and superconductivity. He is a member of the National Academy of Sciences, the American Academy of Arts and Sciences, and the Swiss Physical Society and a Fellow of both the American Physical Society and the American Association for the Advancement of Science. Matthias has received many scientific honors and awards, the two most recent awards being the Oliver E. Buckley Solid State Physics Prize in 1970, and American Physical Society International Prize for New Materials in 1979. The latter was for discovering unusually high temperature superconducting intermetallic compounds and alloys and demonstrating their usefulness in producing high magnetic fields for electric power technology and magnetic confinement of plasma. Originally from Frankfurt, Germany, he earned his Ph.D. from the Swiss Federal Institute of Technology in 1943.

39

molecules within the cell.

Just as molecular biologists must be able to isolate and purify different biochemicals from the complex mixtures collected from disrupted cells, cell biologists must be able to obtain pure populations of cells from a heterogenous tissue or organ. Current techniques for separating viable cells include electrophoresis, centrifugation, and flow sorting. The first two are bulk isolation techniques. The third, when coupled to flow cytometry, can sort individual cells based on the variables measured on a particular cell. Thus flow sorting (Fig. 2) is a more precise method of separating closely related but functionally distinct cell types than either electrophoresis or centrifugation.

Flow cytometers were developed in the 1960s at Los Alamos Scientific Laboratory (LASL) and independently by Gohde in Germany. Although the first instruments lacked resolution, scientists soon recognized their potential for monitoring the growth pattern of cells, the transformation of cells from normal to malignant, and the function of the immune system. Several groups pioneered the early development of flow cytometry and its application to major problems in biomedicine. This extraordinary technique has been applied to problems in cancer diagnosis and treatment and to studies of basic cellular processes in normal and abnormal cells. Among early expectations was the possibility that this technique could be used for automated cancer detection and thus perhaps for mass cancer screening. This possibility still exists, but we must find new measurement variables that more clearly differentiate normal from malignant cells before it can be realized.

Early Staining and Measurement Techniques

The ability to stain DNA and other specific biochemical constituents of cells, the cornerstone of flow cytometry, dates

A

back more than 50 years to the work of Feulgen and Rossenbeck, who developed chemical procedures that allowed stoichiometric staining of DNA, the genetic material in cells. For the first time, the presence of DNA and its localization in the cell nucleus could be seen through a microscope. The procedure, called the Feulgen reaction, has been used widely to locate DNA in its various configurations including its condensation into chromosomes. A modification of the Feulgen reaction. with fluorescent staining, developed in the 1960s, still serves as a standard for other staining procedures.

The first attempt to determine the quantity of DNA in a cell nucleus by optical means was made by Casperson in 1936. He developed a microscope

photometer to measure the amount of light absorbed by DNA. More absorption corresponded to more DNA. The correspondence was not exact, but it was useful nevertheless.

In the 1950s, Barder, Atkins, Mellors, Tolles, and others, using the early microscopic techniques, observed that elevated DNA levels are characteristic of cells derived from a large number and variety of human tumors. Thus the detection of malignant cells and, hence, clinical diagnosis might be based on the recognition of populations of cells with abnormal DNA distributions. However, microscope techniques are very slow and painstaking. Atkins, a prodigious worker in this field, spent years gathering data that can be acquired in minutes with flow cytometry.

The Beginning of Flow Cytometry

When biophysicists at LASL developed the first flow cytometer, they were studying the effects of radiation on cells. The biophysics group had been concerned with monitoring the effects of radiation on whole organisms. How large a dose, they were asking, is required to affect life span measurably or to change tumor incidence? In 1965, the Atomic Energy Commission changed the direction of the biology program to the cellular level. At that time, the primary analytical tool available at LASL for monitoring cells was the Coulter counter, an electronic device that counts cells by measuring changes in electrical resistance. To use the Coulter counter, cells, immersed in a conducting medium, are passed through an insulating orifice. Because biological cells are quite good insulators, they decrease the conductivity across the orifice as they pass through it. The Coulter counter converts the decrease in conductivity to a voltage pulse for counting the number of cells per unit volume.

At that time, physicists who had transferred from nuclear reactor work to biophysics were using the techniques of gamma-ray spectroscopy to detect the presence and character of radioactive materials in humans and animals by counting and measuring the number of gamma rays emitted from an organism. By adapting the techniques of pulseheight analysis to the analysis of voltage pulses from the Coulter counter, they converted the Coulter counter into a device that quantitated cell volume. Now, volume distributions of large populations of cells could be measured. The desire to examine the cells corresponding to a specific volume led to another important development, the design by Fulwyler of the automatic cell sorter. The group used this device to sort individual cells with a specific volume into a separate container.

The volume-sorting instrument soon was applied to monitoring the life cycle of multiplying cells. LASL scientists considered cell volume a useful parameter to measure because a cell's DNA content doubles during the life cycle to insure proper transfer of genetic information when the cell divides, and an

increase in cell volume must accompany the increase in DNA content. Detailed studies led to the conclusion that cell volume is not a unique marker to differentiate cells at different stages of the life cycle. Fortunately, another parameter, DNA content, is unique. In 1966-67 Mullaney and Van Dilla constructed the progenitor of the LASL flow cytometers, an instrument that measures the fluorescence of a single cell as it passes through a laser beam. This device allows us to measure the DNA content of each cell in a population, if the cells have been stained with a dye chemically specific for DNA. The measurement allows us to follow the normal growth of cells or their abnormal growth caused by a perturbation of the cell's environment or as occurs in dis

eases.

Almost immediately, the biologists in the group were interested in using the flow cytometer to analyze the life cycle of exponentially growing cell populations by measuring the DNA distribution in cells exposed to various experimental conditions. The National Cancer Institute saw flow cytometry with its high accuracy and precision in measuring large populations of cells as a possible tool for early diagnosis of cancer, when the frequency of malignant cells is very low. Several groups including the LASL scientists, Wheeless and his group at the University of Rochester, Sweet and Bonner at Stanford University, and others improved the instrumentation and pioneered the application of flow cytometry to cancer diagnosis and to broader studies of cancer and the immune system.

How the Instrument Works

All flow cytometers have three basic components: (1) a flow chamber in which cells are aligned for measurement; (2) a system for optical measurements consisting of a light source (usually a laser), beam-shaping and collection optics, and a light-detection device; and (3) electronics for signal acquisition, analysis, and display. The entire instrument is shown in Fig. 3 and details of the flow chamber, the optics, and the electronics are shown in Figs. 4, 5, and 6, respectively.