Marsden Fund Newsletter
no.9 September 1999

Contents

1. New research grants
2. Underground lasers
3. A use for idle computers
4. News from Marsden Cottage
5. Using bioluminescence to map activity in soil
6. Funding highlights for 1999
7. How kelp survives on sothern New Zealand coasts
8. Mathematical modelling helps understand where gold and silver end up
9. Kiwi super-vaccine helps boost immunity
10. General comments on the 1999 applications

The money is this year's share of a $23m commitment of Government funding to the Marsden Fund. Each successful project receives funding for up to three years.

The biggest grant of the 1999 awards goes to a major Otago-based initiative looking at the workings of mammalian DNA. A team of scientists who last year succeeded in producing a mysterious fifth state of matter known as Bose-Einstein condensate also received funding to continue their pioneering work.

Other successful projects range from investigations into how a child's memory develops to fresh ideas in climate research; from the origins of interstellar dust to the making of New Zealand English; and from studies of molecular architecture to explorations of the navigational skills of fish larvae.

The 74 projects receiving funding have been selected in a two-step process. From a total of 773 applications, 132 were asked to submit a full proposal, which amounted to requests for more than $24,367,000.

The Marsden Fund Committee noted that all full proposals were worthy of funding and that tough decisions had to be made in allocating from the amount available.

The Marsden Fund supports excellent fundamental research, irrespective of topic or science area. Government provides funding for projects that will widen the knowledge base in New Zealand, enhancing the country's ability to participate in, and benefit from, research of an international standard.

The Marsden Fund is a contestable fund administered by the Royal Society of New Zealand. The Fund was named after Sir Ernest Marsden, founding Permanent Secretary of the DSIR in 1926.

For highlights of the 1999 awards, click here.

The Cashmere Cavern is home to the world's most sensitive stand-alone instrument for measuring the Earth's rotation. This million-dollar ring laser equipment is based on supermirrors of such unprecedented quality that it was previously restricted to defence applications.

Beyond their measurement role, the lasers detect so-called teleseismic waves in the Earth's crust and are helping provide a seismic listening post for earthquakes as far away as Papua New Guinea and Vanuatu.

The Canterbury Ring Laser project, regarded by scientists as the most innovative of its kind in the world, was officially opened in 1997 when a new $1.5 million super-sensitive instrument built in Germany was set up in the darkened underground laboratory.

The project is the outcome of an international collaboration between German, American and New Zealand researchers. Installation of the new laser, known as C-II, was supported by the Marsden Fund.

"In January and February 1999, we found that we were able to detect minute rotations from teleseismic waves in the Earth's crust. They were triggered by magnitude 6.5 and 7 earthquakes in locations as far away as New Britain, Papua New Guinea and Vanuatu, i.e., from more than 40 degrees away around the Earth," explained Geoff Stedman, New Zealand project leader and University of Canterbury physics professor.

"So the Cashmere Cavern has become a seismic listening post of novel and international interest. Standard seismometers measure linear motion. The Canterbury instruments are unique in measuring rotation. This is the first time that rotational effects from such remote earthquakes have been detected.

"Not only does the C-II ring laser clearly measure the horizontal teleseismic rotations, but the relatively cheap (if relatively unstable) very large vertical ring laser G0 clearly measures the corresponding vertical rotations. All these devices including a standard linear seismometer are now synchronously logged."

Professor Stedman explained that in December 1998 a major upgrade to the Cashmere facility yielded significant improvements. "The systems were made more symmetrical by rearranging the curved mirrors and by removing mirror mountings. New mirrors were also installed," he said.

"The result is a 100-fold decrease in frequency drifts and a corresponding increase in sensitivity. The frequency-pulling effects of backscatter were reduced dramatically by once again challenging the conventional wisdom for laser gyros and by allowing changes to occur with the effects of pressure."

Geoff Stedman and his colleague Clive Rowe visited Germany in June to contribute to the final design review for the large and expensive German ring laser G, for which the Canterbury lasers are essentially pilot or trial machines.

"This review included ceramic specialists, optical fabrication specialists, engineers, a geologist, and architect. It gave us some fascinating insights and showed the big German project is well on target," he said.

"Our contributions related heavily to our experience on the Canterbury lasers, on such matters as the pressure vessel, the laser alignment, the laboratory environment, the effects of mirror backscatter, thermal conduction and so forth."

Professor Stedman said the laser block for the new German ring laser is a low expansion, oven-top ceramic, Zerodur, which is extremely stable. The block weighs 10 tonnes and was being cut and polished at Schott Mainz. Construction on the new underground laboratory in Bavaria, Germany, would begin during September 1999.

For further information, contact Prof. Geoffrey E Stedman at the Department of Physics & Astronomy, University of Canterbury Phone: (03) 3642549 Fax: (03) 3642469 Email: g.stedman@phys.canterbury.ac.nz Address: Private Bag 4800, Christchurch, New Zealand

German researcher Dr Ulrich Schreiber adjusts the ring laser equipment at Cashmere.

Known as Kalaka, the Linux-based computing system running on 160 PCs is believed to be the most powerful computing device in the Southern Hemisphere.

Creating the supercomputer has allowed Mathematics Professor Marston Conder and one of his PhD students, Peter Dobcsanyi, to develop sophisticated ways to find discrete structures with high degrees of symmetry, specifically arc-transitive graphs and regular maps on surfaces. It is an area of mathematics that lends itself to parallel computing.

According to Dobcsanyi, the computing system was running on a cluster of PCs in student computer laboratories
at the University's Tamaki Campus. He named the system "Kalaka" after a Hungarian village custom where people join forces to perform a task that would otherwise be too difficult to accomplish by a person or a family.

"Building Linux clusters for high performance scientific computing was initiated with the Beowulf projects at NASA (USA) in 1994. Since then many similar systems have been built around the world, but Kalaka is unique in the way that it was set up using idle-time of computers otherwise dedicated to very different tasks," he said.

"Also, the cluster, connecting over 160 Pentium II 350Mhz machines, is one of the largest among similar systems, and to the best of our knowledge, the most powerful computing device in the Southern Hemisphere."

Professor Conder said using the computers during idle-time has necessitated working late nights, weekends and semester breaks. "But the results have been spectacular, significantly pushing back the boundaries of what was previously thought possible from a mathematical point of view.

"To mention a few results, this research has produced a complete list of all trivalent symmetric networks on up to 768 vertices, all regular maps on orientable surfaces of genus up to 15 (including both reflexible and chiral examples), and regular maps on non-orientable surfaces of genus up to 30."

For further information, contact Professor Marston Conder at the School of Mathematics and Information Science, University of Auckland Phone: (09) 3737599 Email: conder@math.auckland.ac.nz

News from Marsden Cottage

by Dr Valda McCann, Manager, Marsden Fund

A time of tough decisions

A total of 74 of the 132 full proposals submitted for 1999 have been funded. This represents a greater proportion than usual and is the direct result of an increase in Government funding of $1m in the 1999/2000 financial year. The panels and the Marsden Fund Committee noted that all proposals submitted at the full proposal round deserved funding but limited financial resources meant difficult decisions were necessary.

Below are tables summarising data on the 1999 applications, by panel. The figures include proposals assessed by more than one panel. Of the preliminary proposals the Committee chose 132 to be invited to submit full proposals and two full proposals were considered by two panels. Of the successful proposals invited to take up contracts, two have been allocated funds from both the Humanities and Social Sciences panels. Applications assessed by more than one panel account for the different totals of proposals in the two tables.

Some of the successful applications are highlighted elsewhere in this issue. The announcement of the successful applicants was accompanied by a media pack, including brief articles about some of the new research.

An intriguing feature of the projects is the collaboration that will occur – more than half of the projects involve researchers at different organisations. This year three technical institutes are involved in Marsden research. Also represented are several other organisations that have not previously been involved in Marsden contracts.

New staff

Meredith O'Brien has left the Royal Society and taken up a new position. She will be much missed for her efficiency and helpfulness. However we have been fortunate to be able to appoint Rochelle Barton as Administration Officer and she will take up the position on September 13. She has an arts degree, a Diploma in Library and Information Studies and a background in administration and editorial work. She joins the Marsden group at a busy time when contracts are being drawn up.

Application forms for 2000

The new application forms will be available from our office and also on the web in early November. Until the new documents are put on the web this year's ones will remain as a guide. Although the general format will be the same, changes will be made to the application forms to improve the information given to the panels and to clarify instructions. The timetable will be similar to this year's with the applications due in early February 2000. The Marsden Committee will finalise these details at a meeting in October.

1999 applications

Panel	Number of preliminary proposals received	Number invited to submit a full proposal	Number successful
Biochemical and Biomedical Sciences	115	17	11
Cellular, Molecular and Physiological Biology	171	25	12
Ecology, Evolution and Behaviour	175	23	12
Earth Sciences and Astronomy	79	13	8
Humanities	37	10	6
Mathematical and Information Sciences	46	13	6
Physical Sciences and Engineering	119	18	12
Social Sciences	92	15	9
Total	834	134	76

Summary

Number of applications at the preliminary round	773
Number of applications invited to submit a full proposal	132
Number of applications invited to take up a contract	74

Marsden at a glance

The Marsden Fund supports excellent research, in a wide range of topics covering the sciences, social sciences, humanities and engineering.

Each year, Government provides funding for projects that will foster research of the highest calibre. This is work not subject to Government priorities but which will nonetheless enhance New Zealand's ability to participate in, and benefit from, research of an international standard. Set up in 1994, the Marsden Fund is a contestable fund administered by the Royal Society of New Zealand.

A Marsden Fund Committee of 10 eminent researchers, chaired by Professor Diana Hill, is appointed by the Minister of Research, Science and Technology to make recommendations for funding. Selection criteria focus on the merit of the proposal, the potential of the researchers to contribute to the advancement of knowledge, and the enhancement of research skills in New Zealand, especially those of emerging researchers.

Eight panels have been established to help the Marsden Fund Committee assess proposals. These are:

• Biochemical and Biomedical Sciences
• Humanities
• Cellular, Molecular and Physiological Biology
• Mathematical and Information Sciences
• Earth Sciences and Astronomy
• Physical Sciences and Engineering
• Ecology, Evolution and Behaviour
• Social Sciences

For more information and application forms, contact the Marsden Fund c/o the Royal Society of New Zealand, 9 Turnbull St, Thorndon, P O Box 598, Wellington, New Zealand.

Phone: (64–4) 4728345Fax: (64–4) 4731409Email: marsden@ rsnz.govt.nz

In a project involving Soil and Earth Sciences and the Molecular Genetics Unit at Massey, Marsden Postdoctoral Scholarship holder Dr Charles Russell is using so-called bioluminescence techniques on soil bacteria to help map activity in soils, particularly in the soil/root zone.

A natural phenomenon often found in the marine environment, this light emission or bioluminescence, is caused by the oxidation of certain large carbon compounds by an enzyme called luciferase.

In the Massey study supported by a Marsden grant, the genes of luciferase have been introduced into soil bacteria using molecular tools called bacterial plasmids. These allow gene transfer from one bacterial type to another.

"This allows us to insert the genes at random into the DNA of a bacterium," Dr Russell explained. "Genetic modification of this type does burden bacteria, often causing detrimental effects that prevent them operating normally.

"Ecological biosensors must be sensitive and produce an intensity of light that can easily be measured in soil while at the same time be able to exist in a natural soil environment."

In the course of the experiment, bacteria from the root surface have been successfully selected as suitable biosensors. This was done using a series of growth and root colonisation studies using a technique called Biolog.

"Biolog allows us to compare the growth characteristics of modified bacteria to their wild type parent strain," Dr Russell added. "Most people knew that forests, crops and pastures all rely on soil bacteria to degrade and recycle nutrients from plant litter and animal excrement.

"Although many of the biochemical processes are quite well understood, how these micro-organisms operate in the soil/root environment is poorly understood," he said.

For further information, contact Dr Charles Russell at Massey University Phone: 06–3569099 Email: c.n.russell@massey.ac.nz

Cultured marine bacteria (Photobacterium fischeri).

The Massey work is showing that bacterial species are very niche specific. This means that different types are needed for different niches when using these tools as indicators of population dynamics in space and time. For example, a root surface specific bacterium will give information on carbon compounds being directly released from the roots. Another bacterium, which occupies a niche further away from the root surface, may use soil organic matter in conjunction with other root carbon compounds.

Grass roots inoculated with bioluminescent root bacteria (left image bright field, right image dark field).

Experiments have shown that the biosensor population in soil is sensitive to different sources of carbon containing compounds moving through a thin soil zone. Luminescence values have been correlated to the activity of normal soil bacteria, and the technique could be used to establish available amounts of carbon compounds in different zones of soil.


Slicing soil into 0.5mm sections.

A group of Otago University physicists has won a 1999 Marsden grant to continue their landmark work on a mysterious fifth state of matter, at laboratory temperatures a billion times lower than deep space. The award is worth $230,000 for the first year, rising to $243,000 for the second and third years.

Project leader Dr Andrew C. Wilson's work on laser cooling of atoms has helped make New Zealand a leading player in this new and rapidly expanding field. His group has a state-of-the-art atom cooling and condensing apparatus, and world-class theoretical and experimental physics expertise.

By cooling atoms at close to zero temperatures, Dr Wilson and his group succeeded in 1998 in reproducing a bizarre and only recently discovered state of matter known as Bose-Einstein condensate. Creation of this long-hypothesised super-atom is being hailed worldwide as the dawn of a new era of atomic physics.

Dr Wilson's research group includes Professor Wes Sandle and Dr Jocelyn Martin from the university's Physics Department. Mr Jan Arlt from Oxford University is also assisting.

According to quantum mechanics, a chunk of matter ripples through space like a light wave.

In the 1960s, light waves were induced to cooperate and form a laser beam, which has proved immensely useful. In a gas of atoms the ripples are normally very confused, and cooperation between atom waves is far more difficult to achieve. By the 1990s, atoms in an ultra-low temperature gas were induced to cooperate (or 'condense'), creating a completely new state of matter, known as Bose-Einstein condensate.

Enormous progress has followed, and atom lasers have been created. Atom lasers are also bright pencil beams of coherent waves, as different from ordinary matter as a laser is different from a candle.

Their study will help the engineering of smaller and faster integrated circuits and will improve interferometers and frequency standards. Dr Wilson's group plans to study how to grow the coherent atomic beam from the normal atom gas.

Along with exploring how best to get the coherent waves out of their birthplace and injected into applications, the group will look at how to make and mix whirlpools in a superfluid atom condensate.

Why aluminium is so toxic to life

A research group headed by Associate Professor Richard Gardner from the University of Auckland has won a 1999 Marsden grant for a project to discover why aluminium is so toxic to plants, animals and microbial cells. The award is worth $155,000 a year for the next three years.

Based at the School of Biological Sciences, Dr Gardner is a pioneer in using yeast as a model to study aluminium tolerance. He believes that aluminium inhibits yeast growth by blocking magnesium uptake – an element vital to all living cells.

The most abundant metal on earth and major component of most soils, aluminium is not essential for life, and can be toxic to plant, animal and microbial cells. It takes a soluble form in acidic soils, which comprise more than a third of the world's arable lands. That means aluminium toxicity is a major factor limiting global agricultural production.

It is also blamed for causing neurotoxicity in mammalian cells and is known to be responsible for dialysis-induced dementias. Aluminium has also been implicated as a factor causing Alzheimer's disease, though hard evidence for this involvement is lacking.

As part of the project, the research group will undertake the molecular analysis of a family of genes that transport magnesium into cells. It will look at how aluminium interacts with cellular uptake of magnesium facilitated by proteins encoded by these genes.

Dr Gardner's research group also includes Dr Peter Ryan from the Division of Plant Industry at the CSIRO in Canberra, Australia, Dr Paul Donaldson from Auckland University's School of Medicine, and Dr Andrew Allan from HortResearch in Auckland.

This work will lead to a better understanding of aluminium toxicity and tolerance in cells, and of magnesium deficiency, a key problem in mammalian nutrition.

Shedding light on grey matter

A group of Auckland physiologists who developed a new technique to shed light on the complexity of grey matter received a 1999 Marsden grant to map brain neurons and show how they process information.

The grant awarded to Dr Gregory Funk, Professor Mark Cannell and Dr C. Soeller at the University of Auckland's Physiology Department is worth $180,000 for the first year, and $170,000 for each of the following two years.

Researchers have long tried to explain the inner workings of the brain but mystery continues to surround exactly how its cells process signals and turn them into actions

The Auckland group uses brain cells that drive respiratory muscles. Like most neurons, these cells have a tree-like structure which branches out into a complex network of dendrites intertwining with those of other neurons.

Each cell is equipped with both specific receptors and synapses, the first to react to chemicals sent out by other cells and the latter to emit these neurotransmitters. Signals received from other cells are transmitted along the nerve cell in the form of currents.

Neurons are often compared to electrical cables, but the researchers are now able to demonstrate that signal transmission in a neuron is more complex than that. With the help of the group's new optical technique, the two-photon molecular excitation microscopy, neurons can be shown in their three-dimensional structure with a detailed map of the distribution of receptors.

In combination with conventional electrophysiological techniques, the researchers will be able to examine the anatomical
structure of nerve cells as well as their behaviour, correlating changes in morphology with the consequent functions.

The group will also develop a new technique that will allow them to excite or block out parts of the neuron selectively. These experiments will clarify the hypothesis that spatial distribution of receptors creates pockets of specialised responses along the dendrite membrane.

The work will therefore provide new insight into neuronal integration and how the brain achieves such remarkable information processing.

Thinking about thinking computers

Associate Professor B. J. Copeland and Diane Proudfoot from the University of Canterbury have received a 1999 Marsden grant to develop logical models for hypercomputers and to review the philosophical principles shaping artificial intelligence.

The Department of Philosophy researchers plan to take a three-pronged approach to their work. They will develop a class of models of hypercomputation and investigate their limits, and describe a new view of the mind, according to which the mind is like a hypercomputer. They will also propose a new set of philosophical principles for artificial intelligence to give this work a new direction. Each of these tasks requires creation of new paradigms.

Successful completion of the project will put New Zealand at the forefront of theoretical research into computation and is likely to bring significant spin-offs for computer science. The group's grant is worth $133,000 a year for three years.

Artificial intelligence, or thinking computers, may be a thing of the future, but the mere possibility that they might exist calls for a fundamental revision of our philosophical thinking about computation.

Hypercomputers can perform tasks customarily described by philosophers and others as impossible to be carried out by machines. If such a machine could be built, it would not only revolutionise computing but also call for a radical rethinking of the philosophical view of the human mind.

In 1934, Alan Turing invented the Universal Turing Machine (UTM), an abstract computing machine consisting of a limitless memory and a scanner that moves back and forth through the memory. His invention led to a set of assumptions such as the belief that the brain's cognitive processing can be perfectly simulated by a UTM and that the mind is some sort of information-processing machine. These assumptions have since influenced the philosophy of computation.

Associate Professor Copeland is the founder of the Turing Archive at the University of Canterbury. His group will set out to review and reject some of the assumptions created in the wake of Turing's invention.

Life-threatening fungus going under microscope

Massey University's Dr Jan Schmid and Dr Richard Cannon from Otago University have won a 1999 Marsden grant to investigate a fungus responsible for life-threatening infections. The group's grant is worth $180,000 a year for the next three years.

The research team will study the genetics of a group of particularly harmful Candida albicans strains to find new methods of prevention and treatment. Dr Peter Lockhart from Massey is also part of the research team.

Candida albicans is a major fungal pathogen in humans, causing both superficial infections such as oral thrush as well as life-threatening infections of the blood stream. In New Zealand the number of reported infections doubles every three years, and the fungus poses an increasingly serious threat to human health worldwide.

Preventing and treating these infections is of great medical importance, but a lack of fundamental knowledge of the genetic differences between the various forms of the fungus has long been a hurdle.

Dr Schmid's group has previously discovered a group of Candida strains that can cause disease as many as a hundred times more often than others. The group will use modern DNA techniques to compare the genetic make-up of these strains with a large number of other Candida albicans types.

The group aims to map the organisms' genomes in order to find the genes that make some strains so successful as pathogens. The experiments will help the researchers to deduce which gene products are responsible for causing human disease. The findings will then be confirmed by examining the genes expressed by strains of Candida albicans obtained from human carriers and patients with oral thrush.

The results of this research will provide the basis for new approaches to the prevention and treatment of fungal infections in humans.

Getting to grips with granular flows

A group of applied mathematicians from Industrial Research Ltd has been awarded a 1999 Marsden grant to construct a new theory to describe the way grains flow. The award is worth $110,000 a year for three years.

Led by Dr Graham Weir and including scientists Drs John Burnell and Shaun Hendy, the IRL research group will construct a theory of granular plasticity to provide fundamental new insights into this complex area.

Material in granular form may act like a solid, liquid or a gas. Granular flows crop up in many industrial and geophysical situations, but their behaviour is not nearly as well understood as ordinary fluids. For example, there is still no way to describe a slow dense flow of grains in vertical chutes or hoppers.

Granular materials display a rich variety of physical behaviour. Ordinary fluids have good models at the microscopic level of individual particles and at the macroscopic level for bulk fluids. But the mechanics of granular interactions are not well understood at either of these levels.

The research group's theory will be developed to help predict the velocities and stresses in quasi-static chute and hopper flows of granular materials.

Experiments show that granular material is drawn slowly down a chute as a piston is lowered. A slow moving boundary layer forms on the walls of the chute, about 10 grains thick.

The key point is that the boundary layer remains 10 grains thick, if the chute is widened or if the size of the grains is changed. So the size of the boundary layer can only be predicted if the particle size is used in the model. Standard continuum models, which contain no information about particle size or granularity, cannot explain this boundary layer.

Finding a theoretical description for granular flows remains an enormous challenge for researchers. Although there is an extremely practical application of this work, it also represents an advancement of knowledge at a fundamental level.

"New Zild" speech: the origin of a unique species of English

A group of Canterbury University linguists has been awarded a 1999 Marsden grant to track the development of New Zealand speech from its origins to the present day.

Associate Professor Elizabeth Gordon from the Linguistics Department at Canterbury University and Dr Margaret Maclagan from the Department of Speech Language Therapy are leading the project. Professor Lyle Campbell from Canterbury University's Department of Linguistics is also part of the group.

The researchers will draw on four archives of data that within them contain evidence of the spoken English of New Zealanders born here from 1850 to 1975.

Although the English-speaking migrants to New Zealand brought with them a wide range of English dialects and accents, the variety of English used in New Zealand is surprisingly uniform and like none of the sources which shaped its development.

The availability of recordings of elderly New Zealanders made earlier this century means that it is possible to trace the growth of New Zealand English virtually from its very beginnings. No other variety of English has such a complete record.

The project will build on the increasing body of work undertaken in recent years on the characteristics of the way English is spoken in New Zealand.

The researchers hope that the project will bring to light important information, not only on the origins of our variety of English, but generally on the formation of new dialects and on language change.The award is worth $140,000 a year for the next three years.

Waikato researchers to probe mangrove mystery

A group of Waikato researchers has been awarded a 1999 Marsden grant for a major ecological project to help solve the mystery of why mangroves restrict themselves to warmer parts of New Zealand.

Associate Professor Allan Green and Professor Warwick Silvester from the University's School of Biological Sciences are leading a project that will combine modern research approaches in plant physiology, genetics and ecology to find out why New Zealand mangroves are only found north of about 38°S.

This boundary assumes even greater importance because, unless planted and carefully nurtured by gardeners, the native kauri also cannot colonise south of this latitude.

'Keystone' plants in one of the most diverse and productive ecosystems on Earth, mangroves also have immense value to mankind as an important 'nursery' for very young fish.

The local variety, along with their associated plants and animals, are under some threat from pollution. But the plants are also a scientific enigma because of the sharp cut-off of their distribution.

Some researchers believe that the plant's frost tolerance is the answer, while others think that a combination of poor dispersal by mangroves and the absence of suitable habitats further south is the solution to the mystery.

This project addresses one of the key questions in ecology: Why are animal and plant species restricted to certain areas? Answering this will help researchers understand, for example, why some biological 'invaders' fail, while others succeed.

The research group also includes Dr Chrissen Gemmill from Waikato University's School of Biological Sciences and John Leathwick from Landcare Research in Hamilton. The award is worth $96,000 a year for the next three years.

Shooting stars could illuminate questions about the universe

Professor Jack Baggaley from the University of Canterbury has been awarded a 1999 Marsden grant to determine the source of the tiny grains of dust seen as shooting stars as they enter Earth's atmosphere. His pioneering work on interstellar dust radar detection could help answer questions about the origin of our solar system.

The trail left by meteors as they plunge into the atmosphere can be detected with the help of radar techniques by bouncing a radio signal off the moving particle and recording the echo. New Zealand is home to a unique radar facility, the Advanced Meteor Orbit Radar or AMOR, which can determine accurate positions for the meteor's trail.

Using this technique, Professor Baggaley has already been able to demonstrate that some of the fiery dust grains piercing the Earth's atmosphere enter at much higher speed and a different angle from most particles.

Most of these dust grains originate in our own solar system as debris left over from the break-up of comets. However, a few enter our atmosphere directly from interstellar space – the vast regions between the stars. These particles form dust clouds in space which are known to be the nurseries for young stars, newly formed planetary systems and perhaps life itself.

The conclusion that, while most of the dust grains come from within the solar system, a small percentage originates in outer space, brought Professor Baggaley's group acclaim in the international science community.

The proposed construction of a second antenna will improve the AMOR system and greatly enhance the group's ability to detect the meteors of interstellar origin and to determine their properties. The group is also working with scientists from the European Space Agency, comparing their data with those obtained from space probes.

Their results will lead to a better understanding of interstellar dust and its role in the formation of planetary systems such as our solar system. The grant is worth $145,000 for the first year, and $135,000 for each of the following two years.

How salt is shunted in and out of cells

Dr Fiona McDonald from Victoria University's School of Biological Sciences has been awarded a 1999 Marsden grant for pioneering studies on sub-microscopic openings in cells vital to living organisms.

Known as sodium ion channels, these tiny pores control such vital processes in the human body as blood pressure, pain perception and taste. Separate sodium and chlorine ion channels move salt (sodium chloride) in and out of cells.

The question of what switches them on and off is of great scientific and medical interest. Researchers know such channels are formed from protein molecules called alpha, beta, gamma and delta. Different tissues have ion channels containing different proportions of them.

Dr McDonald's research will focus on the so-called delta protein found in tissues from the brain, pancreas, testis and ovary. She is particularly interested in what controls the activity of this protein and has discovered a fifth protein, called p80, that appears to be closely involved in this process.

She will use a combination of micro-techniques, including mass spectrometry, to identify the gene that codes for the p80 protein. Once this is known, she will be able to determine the molecular basis of its interaction with the delta protein from the sequence of amino acids in the protein.

From this Dr McDonald will deduce how it controls the action of sodium ion channels in a number of important tissues of the body. The award is worth $120,000 a year for the next three years.

Using plankton to measure ocean turbulence

A group of scientists at the National Institute for Water and Atmospheric Research have been awarded a 1999 Marsden grant to develop a system based on the movement and light adaptation of phytoplankton to accurately describe turbulence of the ocean surface.

Dr P. W. Boyd from the NIWA Centre for Chemical and Physical Oceanography in Dunedin and his colleagues, Dr E. A. Abraham and Dr C. L. Stevens from NIWA at Greta Point, Wellington, will use the microscopic plants as a natural tracer of turbulent mixing of water.

Phytoplankton comprises several types of single-celled algae that drift at the surface of the ocean. The minute plants cannot move actively and rely on ocean currents for transport.

In turbulent waters, phytoplankton is not only moved along the surface but also mixed vertically up and down the water column. It is this vertical mixing that the NIWA group will use to measure turbulence and its variability in a range of ocean locations.

Ocean turbulence has a major impact on the transfer of energy and heat between the Earth's surface and the atmosphere, and is therefore a key factor in the study of climate change. However, turbulence is difficult to measure accurately if one relies on physical properties only.

Like all other plants, phytoplankton lives by photosynthesis, a diet of light and carbon dioxide, to produce oxygen and
carbohydrates. When the cells are pulled further down the water column they have to adjust their metabolism to the change of light which is filtered differently by thicker layers of water. A time lag exists between the vertical movement of the cells and their light adaptation, and measuring the plankton's reaction provides a tool for a time-averaged and hence meaningful estimate of turbulent fluxes.

The researchers propose a model based on both the physical measurements of the ocean surface and the biological reactions of phytoplankton that would better quantify and describe the intermittent nature of turbulence. The group's grant is worth $110,000 a year for three years.

Mining texts for meaning and information

Data mining allows for the fast and accurate search for patterns in sets of data. Text mining takes the same approach, but the unstructured and amorphous nature of text makes it much harder to work with.

Professor Ian Witten and Dr Malika Mahoui-Guerni from the Computer Science Department at the University of Waikato have been awarded a 1999 Marsden grant to develop computerised techniques to automate text mining.

In modern culture, text is the most common form of information exchange. Text mining, or the automated extraction of information for particular purposes, is a compelling yet difficult option for speeding up the transfer and extraction of relevant information.

Potential applications for the technology abound. For example, automatic extraction of titles and authors' names is important for the construction of bibliographies and the maintenance of digital libraries. Identification of data from formatted tables could assist in the creation of databases such as stock-market information. Identification of names would allow for the automatic insertion of hyperlinks to other places referring to the same name.

Analysis of natural language is commonly thought of as a problem for artificial intelligence, especially since extravagant attempts at mechanical translation in the 1960s failed.

However, the new approach taken by the researchers is based on the hypothesis that you don't need to understand a text in order to extract useful information from it. Different kinds of data can be recognised on the basis of how well they are compressed by various procedures.

The group supported by this Marsden grant has an international reputation in text compression and machine learning, and the researchers use dynamic-programming algorithms to recover underlying patterns in any document.

They propose a system that users will be able to train to recognise different types of items (e.g., names, dates, titles) without any programming knowledge. The group's grant is worth $88,000 a year for three years.

Southern New Zealand's rocky shoreline, however, hosts some of the biggest and most fascinating seaweeds in the world – with individual kelp plants weighing in at over 70kg, with as many as 200 individual leaf-like fronds!

The mystery of how these plants survive – and thrive – on this 'high-energy' coastline is the subject of a Marsden project involving researchers Craig Stevens from the National Institute for Water and Atmospheric Research at Wellington's Greta Point and Catriona Hurd from the Botany Department at Otago University.

For the past year, the pair have been observing kelps in their own environment, exploring how they evolved and how they survive.

The initial phase of their field programme involved a range of new measurement techniques, including accelerometers and displacement devices attached to the kelp. These are providing a new picture of plant survival in the rocky wave-swept intertidal environment.

"Our focus is on the two largest kelp species in New Zealand, Durvillaea willana and D. antarctica. The blades of the latter are honeycombed with air-filled spaces and so are very flexible, elastic and float. The former, on the other hand, does not float and is comparatively rigid," explains Stevens.

"It is fascinating that these two fundamentally different but closely related species are often found in the same location. They represent different evolutionary solutions to the same environmental problem."

Stevens says accelerometers and displacement devices have been applied in ways never conceived by their manufacturers. The equipment literally takes a pounding – running repairs are the name of the game for the field work.

"The accelerometers were embedded within the kelp blades whilst the displacement devices were connected between the kelp base and a location on the rocky sea floor. All this happens a metre under water with waves crashing all around."

A nearby wave-gauge that measures water height variations, provides a measure of the incident waves. Because the response to the waves occurs in less than a second, synchronisation of all data is the key to this observational approach.

"Already we have been able to determine that our newly developed methods are viable. Further to our expectations, the accelerometer data enable us to map the evolution of individual kelp blade orientation. In addition, we can simultaneously measure the arrival and velocity of a breaking wave with the displacement of the kelp base and the subsequent tossing and turning of its fronds," says Stevens.

The results for the two different kelp species suggest that the elasticity of Durvillaea antarctica is the key to its survival. If the less stretchy Durvillaea willana were also to float at the surface it would be subject to much higher accelerations and subsequent forces.

Further work will target the interaction between fronds as they twist and combine to form what are effectively 'super-fronds', able to withstand the largest of waves.

For further information, contact Dr Craig Stevens at the National Institute for Water and Atmospheric Research. Phone: (04) 3860300 Email: c.stevens@niwa.cri.nz Address: PO Box 14–901 Kilbirnie, Wellington

A section of Durvillaea antarctica blade. The hole in the top piece was the location for an accelerometer.

Tracking the gold in them thar hills

Scientists and miners have long been baffled at how deposits of prized metals like gold and silver move around beneath the earth's surface.

A team of Marsden researchers from Industrial Research Ltd are now building mathematical models to help understand how these precious ores are squeezed between rocks and carried about by water in a dissolved form.

The blazing temperatures and aggressive chemical conditions found in New Zealand's geothermal areas are providing them with a natural high temperature laboratory in which to do their work.

"As you can imagine, reactions between groundwater and rocks are incredibly slow, weathering of rock may take many hundreds of thousands of years. However, at high temperatures these reactions take place very much faster," explains Dr Stephen White, one of the researchers.

"New Zealand is fortunate in having the Taupo Volcanic Zone where the high temperatures and aggressive chemical conditions make for interesting geochemistry and geology."

According to White, the aim of the project is to understand the processes associated with the reaction and transport of chemical species in the earth well enough to model them mathematically, and so discover the conditions that lead to localised deposition of precious minerals.

"These models are being applied to the processes associated with ore deposition, particularly that associated with geothermal fields.

"While the earth may appear to be composed of inert dirt and rocks, it is in a state of continual change as subtle chemical reactions take place between the rocks making up the earth and the water contained in them," he says.

"These chemical processes cause the weathering of exposed rock and create the ore deposits. To understand the changes, we must understand and model both the chemistry of the water-rock interaction and the flow of water within the earth that transports the dissolved chemicals."

White says the most obvious example of these processes is the white veins of quartz often visible in stones and boulders in a riverbed. These are created when quartz from rocks is dissolved in hot water within the earth. This flows through small fractures in the rocks and deposits in the veins as the fluid cools.

Ore deposition occurs when some unusual geological process has acted to separate valuable minerals from rock, transport these minerals and then preferentially deposit them in a small zone.

For example, gold is present in geothermal waters, dissolved from the minute quantities in the rocks. This gold becomes soluble when it reacts with sulfur in the water.

This gold (along with silica and other minerals) is transported towards the surface until changed conditions such as boiling, or reaction with oxygen, cause the gold to deposit.

Often the conditions that cause deposition may be very localised and give rise to rich deposits such as the rich quartz veins found at Golden Cross Mine in Coromandel.

"Developing and solving numerical models of ore deposition is a challenging task," he says. "To treat the geochemistry in a reasonably realistic way may require keeping track of 50 or more dissolved chemical species, several gases, 15 or more minerals and energy.

"We do this by describing the system being modelled using a system of conservation equations. Much effort has gone into developing numerical techniques that allow these equations to be solved," says White.

"We recently took part in a collaborative study with Lawrence Berkeley Laboratory in the United States on the mechanisms for enrichment of copper ore. In this study we modelled the weathering of rocks containing pyrite, one of the most common naturally occurring minerals present in many subsurface environments."

White says pyrite plays a key role in the creation of enriched ore deposits through weathering reactions. It is the most abundant sulfide mineral in many mine tailings and the main source of acid drainage from mines and waste rock piles.

The pyrite oxidation reaction provides a model for oxidative weathering processes with broad significance for geoscientific, geological engineering, and environmental applications.

For further information, contact Stephen P. White or Graham Weir at Industrial Research Limited, Lower Hutt Phone: (04) 569 0000 Email: s.white@irl.cri.nz Address: PO Box 31–310, Lower Hutt

A silica vein through rock containing pyrite is pictured. The darker patches on the right hand vertical surface are pyrite. On the horizontal surface the pockmarks are areas where pyrite has been leached from the rock. The white area is a quartz vein and the dark substance on the quartz is pyrite transported by fluid flowing in the vein.

Fig. 1 Total dissolved sulfur contained in the groundwater as a result of leaching of pyrite- containing rock by oxygen-saturated rainwater.

In a Marsden-funded research project, Drs Margaret Baird and Glenn Buchan from the Department of Microbiology plan to conduct further tests on their so-called 'super subunit' vaccine that is boosting the power of the immune system to protect itself.

They hope their fresh approach to vaccine development may help address some of the long time challenges in the field, including illnesses caused by microbes hidden and multiplying within host cells, and immune system problems that leave some vaccinated individuals unprotected.

"Vaccines are either made up of micro-oganisms that have been killed or rendered harmless, or fragments (subunits) of microbes. We know that subunit vaccines cause fewer side effects but they are sometimes less effective, requiring regular booster immunisations," explained Dr Baird.

"When vaccines are administered, the host's immune system responds by producing antibodies to destroy pathogens that multiply outside cells and killer cells to destroy pathogens that multiply inside them. This response requires the simultaneous production of immune hormones called cytokines. Sometimes insufficient amounts of these cytokines are produced or those that are manufactured drive the wrong type of immune response.

"We are examining ways of overcoming these defects by immunising with a new vaccine made up of a fragment of a micro-organism, a peptide from influenza virus, fused to a specific cytokine. We are testing this by adding it to cultures of immune cells taken from animals. This procedure allows us to model some of the early events that occur when a vaccine is administered to an individual."

According to Dr Baird, immune response to a vaccine can be broadly divided into three stages:

• specialised 'antigen-presenting cells' pick up the vaccine,
• these cells interact with other cells (T cells) to provoke the production of cytokines,
• these cytokines help immune cells to manufacture antibody or killer cells.

"We are using our model to assess the first two stages. The results we have obtained are very promising. We have shown that if influenza peptide is coupled to one cytokine (called interleukin-2) it can turn on T cells much more effectively than just the peptide alone. Furthermore, if we direct the peptide/cytokine product to a particular type of 'antigen presenting cell', we can increase the response dramatically," Dr Baird continued.

"Naturally, we now need to test this effect further. Dissecting out immune responses using culture systems in plastic dishes is rather like using a single violin to convey the melody of an orchestral symphony – it's lovely as far as it goes. But when the complexities of interacting instruments are introduced, quite different effects may emerge."

For further information, contact:
Drs Margaret Baird and Glenn Buchan in the Department of Microbiology, University of Otago,
Phone: (03) 4797285
Email: margaret.baird@stonebow.otago.ac.nz
Phone: (03) 4797708
Email: glen.buchan@stonebow.otago.ac.nz
Address: PO Box 56, Dunedin

Marsden Committee: comments on the full proposals

The marsden committee's general opinion of the 1999 applications is that the breadth of research expertise within New Zealand is highly impressive. All full proposals were viewed favourably and deserved funding but decisions had to be based on the resources available. Requests for first year funding stood at more than twice the monies available. The panellists found that both the referees' comments and the applicants' rebuttals formed a very valuable part of the assessment process. The Committee acknowledged the need to support younger scientists but also to support highly productive established groups. The final recommended proposals include both of these groups.

Though the general comments on quality apply to all panels, specific comments can be made about successful proposals in particular areas. The Mathematics and Information Sciences projects, for example, included theoretical and practical computer science, combinatorics, stochastic optimisation and applied mathematics. This is in contrast with last year when more pure mathematics was funded. In Earth Sciences and Astronomy some excellent interdisciplinary projects are using techniques from the physical sciences and biology. The range of questions addressed in the Social Sciences area underlines the extensive involvement of New Zealand social science researchers in international debates; their research will further strengthen our local understanding and also contribute to global knowledge.