Go to National Library of New Zealand Te Puna Mātauranga o Aotearoa
Volume 77, 1948-49
This text is also available in PDF
(756 KB) Opens in new window
– 100 –

Abstracts Of Papers

A. Wilson Cloud Chamber.

A description was given, with a demonstration, of a diaphragm type of Wilson cloud chamber using alpha particles.

Acoustical Analysis by Variable Density Sound Film.

A description was given of an optical apparatus which the author had devised, which gives the Fourier transform of any ordinary function, provided this function is presented in the form of a light-variation or density-variation such as is the case with sound films. Such Fourier analyses are important in a large number of physical problems, and it was pointed out that in the case of sound waves recorded on films the analysis, is of special interest in that it gives the acoustic spectrum of the original sound. The apparatus performs the analysis almost instantaneously (within 05 second), so that a record may be fed continuously through the machine and the changing character of the spectrum observed visually or photographed on a second moving film. A description was given of the essential features of the apparatus, namely, (1) a rotating glass disc of special construction, having upon its surface a grid of lines or fringes with sinusoidal density-variation, (2) a fine slit which limits the light transmitted by the disc to a narrow diametral portion, (3) an optical system for projecting an image of this slit on to the sound-track or vice versa, (4) a photoelectric cell to receive the resulting light signals, (5) an audio amplifier to amplify them and transmit them to an oscillograph, and (6) an arrangement to provide the oscillograph with a sinusoidal time base, synchronous with the rotation of the disc. Theory was given to show that the oscillograph trace then constitutes the analysis which is required.

The author explained that after the apparatus had been designed and was in the experimental stage, the same method was described by Born, Fürth, and Pringle (Nature, p. 756, 1045), who had evolved the method independently and had applied it successfully to the analysis of a variety of mathematical functions. The above authors used amplitude-modulation of the oscillograph trace, but intensity-modulation was shown to be more appropriate in the present investigation, in order to permit of a continuous or running analysis.

While such running analyses of vibrations would find a useful application in the study of a variety of physical phenomena (mechanical, electrical, acoustical, etc.), the promising field of research offered by speech sounds had been a big incentive in the development of the method. Following on a description of the results obtained with standard wave-forms, which were used as tests of the apparatus, photographs were shown which had been obtained from actual speech sounds recorded on variable-density sound film. The changing nature of the acoustic spectrum, as the vowel quality and the inflection of the voice changed, produced a photographic pattern which enabled certain phenomena of speech production to be studied. A number of harmonic frequencies could be detected, evidently having as their fundamental, at any particular instant, the frequency of vibration of the vocal cords. Changes in vowel quality could be seen to correspond to changes in the relative intensity of the various harmonics, because of variations of the vocal resonances in the cavities formed by the throat, tongue, lips, etc., during articulation. The probable value of these characteristic and readily intelligible patterns in connection with phonetic and related studies was discussed. However, it was not possible to claim complete novelty for the patterns, as during the course of the research a group of collected papers had appeared (Journ. Acoustical Soc. of America, 18, 1, July, 1946) describing a large-scale investigation by a team of workers at the Bell Telephone Laboratories, New York. Here an alternative method, using magnetic tape recording, had been evolved for producing speech patterns, examples of which were shown and discussed. These workers were thus the first to achieve success in what is referred to in one of the papers as the long search for a legible and quantitative display of speech.

In conclusion, the author wished to acknowledge the invaluable assistance of Messrs. C. F. Coleman and J. W. Lyttelton in carrying out the research.

– 101 –

A G-M Counter Used in Prospecting Bores.

A portable Geiger-Muller counter has been designed for remote operation of the G-M tube through 2,000 feet of cable. It has been used to measure gamma lay intensities at various depths in prospecting bores, and cosmic ray intensities in deep water.

Some Consequences of a Modified Equation of State.
By D. B. Macleod, Canterbury University College.

A modification of van der Waals' equation was suggested, assuming the volume of the molecule to be a function of pressure. Various consequences of this were examined and compared with experimental results. In particular, a possible explanation of liquid helium II was discussed in the light of this theory.

The paper summarised five articles recently published in the Transactions of the Faraday Society. (1) Van der Waals' Equation of State and the Compressibility of Molecules. XL, (10), 1944, 439–47. (2) A Calculation of the Latent Heat of Vaporisation based on a Revised Equation of State. XLI (3), 1945, 122–6. (3) On the Direct Calculation of the Viscosity of a Liquid, both under Ordinary and High Pressures, on the Basis of a New Equation of State. XLI (11, 12), 1945, 771–7. (4) On Some Theoretical Consequences of a Revised Equation of State and a Possible Explanation of Liquid Helium II. XLII (6, 7), 1946, 465–8. (5) A New Explanation of Liquid Helium II (with H. S. Yubsley). XLII (9, 10), 1946, 601–16.

Principles of Molecular Distillation.

A vacuum still with the distance from the distilling surface to the condenser reduced to a minimum is termed a “molecular still.” Molecular distillation permits the fractionation of mixtures of heavy long-chain molecules (such as liver oils) without decomposition. Apparatus for such work was described and some of the design difficulties discussed.

A full description of an industrial centrifugal still and equipment will shortly be published in the N.Z. Journal of Science and Technology.

Instrument Testing and Development.

A brief survey was given of the activities of the Instrument Testing and Development Section of the Dominion Physical Laboratory, together with a description of methods and apparatus used. Illustrative examples were shown.

A. New Optical Polishing Abrasive.

An account was presented of the various types of abrasive materials used on optical glass, with special reference to the use of titanium oxide as a polishing abrasive.

Analytical Balances and Their Faults.

A description was given of common faults in analytical balances, together with methods of correcting them. A brief outline of the requirements of reliable instruments was supplemented by a description of suitable mountings for them.

The Present Stage of the Solar Cycle.
By I. L. Thomsen, Carter Observatory.

The paper reported New Zealand observations showing the present trends of solar activity, and indicating what region of the upper part of the solar cyclemay have been reached at the present time. A comparison was made with longterm predictions and the method of indicating the solar activity was described.

– 102 –

Cosmic Rays.
By E. Marsden, Department of Scientific and Industrial Research.

A summary was given of recent advances abroad, especially in regard to latitude variation and altitude variation and to the new photographic technique. Also a programme of observations in New Zealand, and possibly the Antarctic, was submitted for discussion.

Lunar Short-Wave Radiation.

Evidence was presented indicating that radiation from the moon contributes to the ionisation of the upper atmosphere. Analysis of observations of long-distance radio transmission show periodic maxima corresponding with the period of synodical revolution of the moon.

Notes on the Aurora Australis.
By I. L. Thompson, Carter Observatory.

Before 1933 very little exact knowledge was available concerning the appearance of the Aurora Australia as seen from New Zealand. Up to that date most of our knowledge of the Aurora Australis had been obtained from the reports of various Antarctic expeditions, commencing with Cook in 1773, which were of necessity spasmodic and based on short periods of one or a few years. Moreover, it is worth noting that, by coincidence, most of these expeditions took place in the period of the solar cycle from one and a-half years after maximum to the minimum period.

The work commenced by the late Mr. M. Geddes in 1933 and at present being continued as far as possible by the Carter Observatory has given us records of auroral displays over one and a-half sunspot cycles, which by virtue of this long period should provide results of considerable interest. The time has now arrived when this recorded data should be reviewed and analysed. It is considered that each New Zealand aurora recorded has some significance for the polar auroral activity as a whole, because in each case it indicates a northward advance of the zone, and thus an increase above what might be termed “quiet” auroral periods.

Perhaps the most fundamental general result appearing from the work to date is that, as far as can be seen, the southern aurorae are similar in form and height to the northern aurorae. Despite this, the many statements by travellers to the effect that although the same forms may be apparent, the northern aurorae have a quality lacking in the southern aurorae, would indicate the desirability of a direct comparison by the same observer. The same sequence of forms appears to take place as in the northern hemisphere, and several of the remarkable auroral forms studied by Stormer have also been clearly recognised.

The great bulk of the New Zealand work, however, has been of the purely visual type without the use of instruments and there are therefore no data available concerning auroral heights remotely comparable with those of the northern hemisphere. Geddes, working under great difficulties in his spare time, has published results of measures from 18 duplicate photographs. While these indicate that the auroral forms are of the same order of height as in the northern hemisphere, they are not sufficient to provide conclusive data. The cameras were lent to New Zealand by Professor Stormer and are still here. No work has been done since the time of Geddes, and the present policy is that unless a good base-line can be established, numerous photographs of good quality obtained and their rapid measurement and reduction undertaken, it would be better not to continue the work until these conditions can in good measure be fulfilled. An attempt is being made to institute a programme of auroral height measurements.

Another feature of the Aurora Australis appears to be the rather large expansion and contraction of the auroral zone in sympathy with the solar cycle, compared to the northern auroral zone. This is shown to explain the discrepancy which appeared to exist between the observation of Mawson at Macquarie Island at sunspot minimum that the aurorae were seen in the southern sky, and the New Zealand observations of Geddes at sunspot maximum that the auroral forms were either directly over or to the north of Auckland, Campbell, and Macquarie

– 103 –

Islands. It has been further confirmed by recent observations at Campbell Island where aurorae were confined to the southern sky during the last sunspot minimum period, but are now beginning to appear in the northern sky at a time close to sunspot maximum. The study of geographic position of southern auroral forms would therefore appear to be of some importance.

No work, as far as is known, has been carried out in the southern hemisphere on auroral spectroscopy or photometry.

At present, auroral studies are based on numerous reports sent in to the Carter Observatory by voluntary observers scattered all over New Zealand, and from Tasmania and Australia. The establishment of a scientific party on Campbell Island which makes auroral observations is a most valuable aid, and its importance cannot be too strongly stressed. Valuable co-operation exists between the Magnetic Observatory, Christchurch, and the Carter Observatory. The outstanding need at the present moment is for additional staff for the purpose of assisting in the review of material at present on hand and the better organisation of work for the future.

Very Soft X-rays.

A brief description was given of the development of the technique of soft X-ray spectroscopy. This region of the spectrum is distinguished by the extreme absorbability of the radiations and by the fact that the wave-lengths are usually too great to be examined by the normal procedure of crystal spectroscopy. It may be considered to embrace all X-ray wave-lengths greater than about 10Å.

The early investigations were mostly performed by a difficult photo-electric critical-potential technique. Whilst this showed the presence of these soft radiations, and whilst it yielded results of some value, many of the effects observed were difficult of interpretation. It was superseded by the plane-grating method, in which the radiations were allowed to fall on the grating at almost grazing incidence. At very small glancing angles the radiation is totally reflected owing to the fact that the refractive index of any material is, for X-rays, leas than unity. Hence adequate intensity may be obtained from the grating. Fortunately, also, the grazing-incidence position leads to great dispersion and compensates in some measure for the coarseness of ruled (as opposed to crystal) gratings. Finally, provided that it is only a millimetre or so wide, the grating at grazing incidence is self-focusing. It was shown by Compton and Doan that ordinary X-ray spectra could be obtained from a grating used in this way; and by Osgood and Thibaud it was shown to be the ideal means of investigating the soft X-ray region. (Slides of a vacuum spectrometer designed by L. P. Chalklin and the writer were shown, together with illustrations of the spectra obtained.) It was now possible to measure the emission lines and absorption edges of the soft X-ray region and so to check up the values for those energy levels lying near the “surface” of the atom. In the course of this work it was clearly demonstrated that it was possible to obtain radiations due to transitions between energy levels of the same electronic “shell,” e.g., the Miii—Miv, v radiation of molybdenum.

The plane-grating method suffered from the limitation of resolution and of intensity imposed by the small width and consequent small number of rulings of the grating. By employing a concave grating it was possible to maintain the focusing properties and at the same time to increase the size and the resolution of the instrument. All modern soft X-ray spectroscopy is performed with, such gratings. The slit and the photographic plate are usually placed on the Rowland circle of the grating, but owing to the necessity for grazing incidence and the small angles of diffraction, the slit, grating and plate are close together, and the radiation strikes the plate at a very oblique angle. (A slide was shown of a concave grating instrument designed by S. S. Watts and the writer. This instrument was planned to have the merit of being adjustable by optical means without the necessity for tedious trial and error adjustments in which each test would mean a separate sealing and evacuation of the apparatus. A slide of a vacuum spark spectrum showed, by the resolution of close sharp lines, that the theoretical resolving power of the instrument was in fact attained.) In soft X-ray, as in ordinary X-ray spectroscopy, Siegbahn and his co-workers have played a predominant part.

– 104 –

The soft X-ray technique affords the best method of investigating transitions in which valence electrons move to levels within the atom. Such transitions always give X-ray lines of appreciable width. Measurement of this width allows the calculation of the energy spread of the valence electrons. It has been measured for many metals by O'Bryan and Skinner, and in most cases it has been found to be in remarkably good agreement with the simple Sommerfeld theory of metals. (A slide was shown to illustrate the wide emission bands obtained, by O'Bryan and Skinner.)

In the more detailed theory of solids the levels at the “surface” of the atom are replaced in the solid by bands of levels. Neglecting the small effect of thermal agitation, it may be said that the valence electrons of the solid fill the lowest of these close levels, but, by the Pauli principle, only one electron can live in each. Hence all the lowest levels are full and all the higher ones are empty. This sudden discontinuity between full and empty levels should give a sharp short-wave edge to the emission band, and this is, in fact, always found in the case of metals. In the case of insulators, however, it is believed that the valence electrons lie in completed bands of levels and that the lowest empty band of levels is an appreciable energy step above. Application of an electric field cannot in such case accelerate the electrons because this would imply a small increase in their energy and there are no empty energy levels of the required value available. In short, an insulator should be characterised by a full valence band and there should be no sudden division between filled and unfilled levels and no sharp short-wave edge to the soft X-ray-emission band. In some recent work, in which a new method of measuring the intensities in the bands has been developed by the writer, the K-emission band of the insulator diamond has been examined. Although it is not symmetrical in form (and is not actually expected to be) it shows very definitely that there is no sharp edge. (A slide of the diamond curve was shown.)

Until the intensity distribution in the emission bands has been studied thoroughly by the soft X-ray techniques and until the results have received adequate theoretical interpretation, the behaviour of the valence electrons in solids will not be fully understood. The urgency of such investigations is manifest from the fact that the valence electrons govern the conductivity and the cohesion of metals. In conclusion, it may be pointed out that now, in addition to atomic-spectra and band spectra, we have solid spectra and that the work on this subject is in its infancy.

Curve-fitting by Least Squares.

The elementary methods of least squares as applied to curve fitting, including the method of orthogonal polynomials, was discussed.

Powers Punched Card Equipment and Computational Problems.

The application of Powers-Samas punched card equipment to the solution of the problems mentioned in the previous paper was demonstrated on a 21-column Powers equipment.

General Principles of Radar Design.
(Notes from Lecture)
By D. M. Hall, Dominion Physical Laboratory.

At the beginning of 1939 the radar sets used in England were almost exclusively those which operated on approximately 7 metres and were known, as C.H. (Chain Home) sets. These sets had flxed aerial systems and were used solely for detecting high-flying aircraft. The design of these sets was vastly different from those used on shipping these days to pick up marker buoys at 50 yards or the type of set used automatically to track flying bombs.

I will now mention briefly the fundamental factors which determine the-design of a set once its purpose and specifications are known. In the design of these sets one of the most important factors used is that of R.F. frequency of the transmitted pulse. In the earliest sets the maximum frequency available from transmitting tubes with suitable power was approximately 30 megacycles.

– 105 –

Now it is possible to produce more than 200 k.w. on 10,000 mc/s. The need for higher frequencies was made very evident in the early days of the war. This was so because the higher the frequency for a given size of aerial the narrower the beam becomes and for radar sets near sea-level it is possible to illuminate the surface of the sea for a greater distance. This was essential to pick up low flying: aircraft, shipping, submarines, etc. With a narrower beam it is possible, too, to obtain better resolution. In these days it is usual for marine radar sets, air-borne radar sets, and some land-based ones to operate on approximately 10,000 mc/s. Although higher frequencies have been used, there is another factor which comes into play when frequencies on or above this are used, i.e., the metecorological effect. It is evident from the above that for a set to be used in all weather s in tropical regions for long ranges 3000 mc/a, or 10 cm. appears to be about the limit. However, this feature has its benefits as well as detrimental effects, since clouds containing a high moisture content, besides attenuating radio waves also reflect them. This means that with 3 cm. sets it is possible to track thunderstorms for distances of 100 miles or more and forecast with accuracy when they will be overhead. Lower frequencies such as 100 mc/s. are by no means out of date, as these frequencies can still be used to pick up targets at long range that are not near the horizon. A frequency of 75 mc/s. was used to pick up the moon. It is possible at the lower frequencies to use triode valves and hence to use a long pulse length, in the nature of 10μ sec. With this long pulse length the receiver requires a relatively narrow band pass, and in addition it is possible to use R.F. stages of high gain in the receiver.

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

Pulse Length and Shape.—The pulse issued is the main factor in determining the maximum and minimum range of a set. Short pulses are essential if a short minimum range is to be achieved, i.e., 1μ sec. pulse if ranges of 164 yards or more are to be received. 1/10μ sec. pulse if ranges of less than 50 yards. The pulse length also determines the distance between two targets in line it is possible to separate; in other words, the pulse duration must not cover the two targets. If it does, then the two targets merge into one. When a greater range is required, then the pulse length is increased, as this enables a narrow band receiver to be used. For triodes this may be as long as 10μ sec., but for magnetrons the upper limit is approximately 3μ sec., due to moding. Plate modulation is generally used for triodes, and cathode modulation for magnetrons, and in the latter case a negative pulse is applied. Generally, a rectangular-shaped pulse is required for magnetrons with a 5 per cent, rise and fall and no voltage spikes.

Pulse Recurrence Frequency—This ranges from 50 c/s. to 5000 c/s., depending on the type of set used. First, the repetition frequency must be kept low enough to enable the echoes from targets to be received on one sweep; i.e., a P.R.F. of 1000 c/s. enables echoes up to 93 miles to be obtained. If echoes up to 186 miles are to be received, then the P.R.F. must not be more than 500 c/s. The P.R.F. is kept as high as the ratings of the modulator system allows to enable the maximum number of pulses to be received from a target when a beam is sweeping past it. This factor also helps to eliminate fading, which may occur with successive pulses and also increases intensity on intensity-modulated display tubes.

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

Peak Power.—In describing the power output of a radar transmitter two terms are used. (1) Peak power, which is the average power during a pulse (3) Average power, which is the average over the repetition period. Peak powers from a kilowatt to 5 megawatts have been used in transmitting equipment and as much as 250 k.w. are generated by 3 cm. magnetrons. However, although the peak-power output may be very high, the average power output is small because of the difference between the pulse length rr, and the pulse interval σ Peak power/Average power=σ/rr

The maximum range is proportional to the fourth root of the power output, hence to double the range of the set the power must be increased 16 times.

Beam Width of Aerial System..—The gain of the antenna will also affect the range of a radar set. Gain and area are related in terms of wave-length as follows:—

– 106 –

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

G = K A/λ2

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

Where K is a constant of proportionality, λ is the wave-length at which the antenna operates. Therefore, for a given wave-length and diameter, maximum range is proportional to √A. The beam-width is given by θ ∝ Kλ/D

Increasing the antenna area has the additional effect of decreasing the beam angle of the antenna. For aircraft and shipping the aerial size is, of course, limited, and most shipborne radar sets use an aerial of no more than 4 ft. diameter. On 3 cm. this gives a beam of approximately 2 degrees to half power. For the navy and also on bombers to provide for continuous illumination of the target during evasive tactics, gyro-stabilised platforms were used. Other types of aerials used, besides the parabolic type, include slotted wave-guides and for accurate direction finding, nutating dipoles are used and these also enable the radar set automatically to follow the target.

Rates of Angular Rotation.—Generally, for land-based sets a speed of 6 r.p.m. is used, but where information is required, say, by a ship moving up a channel, then higher speeds of rotation are used which are in the vicinity of 60 r.p.m. Scan rates are made as slow as factors permit to allow an increased number of pulses to illuminate the target. Halving the scan rate produces a system gain of 1.5 db. Decay time of the cathode-ray display screen must be closely related to scan rate if all advantages of proper scan rate are to be realised.

Other Factors Affecting Minimum Range.—If echoes are to be obtained immediately following the transmitter pulse, then several other steps must be taken.

(1)

A fast recovery in TR tube. This depends on type of gas used.

(2)

Receiver must not paralyse, hence the use of a pulse during transmitter pulse, applied to the suppressors of two of the I.F. tubes to reduce the gain of the receiver.

(3)

Low time constants in components in receiver.

Required Accuracy in Bearing.—Aircraft and marine radars do not usually give more accurate bearings than 1°. If greater accuracy than this is required, then either: (a) Lobe switching or (b) a narrow beam must be employed, and this necessitates a large aerial.

Accuracy in Range.—A general-purpose search radar is made accurate to a fraction of a mile. For gunnery work and other uses a range accuracy of a matter of yards can be measured. In radar this accuracy remains constant as long as the target is visible on the screen.

Weight of Equipment.—In airborne equipment, engine-driven 800 c/s. alternators are used, to save weight and space, as this cuts down the size of transformers and the amount of filtering required. Shipborne sets use 50 c/s. up to 500 c/s. to save space.

When a radar set is required for a specific purpose, the above-mentioned factors are the main ones to be taken into consideration; but time is too short to deal with many others such as transmission lines, aerial turning mechanisms, etc. The majority of sets built these days use centimeter wave-lengths to enable high definition to be obtained.

Radar and Radio Methods of Position-fixing and Navigation.

Systems providing assistance to the navigation of ships, and employing techniques developed during the war are briefly reviewed.

Radar.—Marine navigational radar is a simple P.P.I. radar designed for high discrimination and good minimum range. An example developed in Britain provides a beam-width of 2 ½ degrees and minimum range and range discrimination each of 50 yd. The latter figure implies a pulse length of ¼ microsecond, and

– 107 –

a receiver with a band-width of at least 6 mc/s. and incorporating special measures to prevent “blocking” by the transmitter pulse. P.P.I. scales giving edge ranges from 3,000 to 80,000 yd. make displays available which are suitable for navigation of close waters, for coasting, or for making a landfall. Demonstrations in Britain have shown that navigation in busy waters in the worst visibility is feasible. This is much facilitated by a display method which provides continuous visual comparison between chart and P.P.I.

The wave-length employed in the radar described above is 3 cm.; at a wave-length of 10 cm., less interference is experienced from heavy tropical rain, but the same discrimination and detection of low-lying land is not possible.

At a wave-length of 3 cm., an aerial 5 ft. wide is needed to give a 2 ½ degree beam-width. The British practice is to rotate the aerial at over 20 r.p.m. in order to obtain information quickly; American practice prefers a slower rate. The aerial must be mounted high enough and far enough forward in the ship to avoid serious masking on ahead bearings.

The pulse repetition frequency should be high, in order to ensure adequate brightness of the display, but too high a frequency may lead to the appearance of long-range echoes on the trace subsequent to that to which they belong. The limit is about 1500 p.p.s.

Hyperbolic Systems.—The group of position-fixing systems known as the hyperbolic systems depend on the measurement of range differences from three or more known points. The range differences may be measured as the intervals between the arrival times of pulses, as in Gee and Loran, or as the differences in phase between CW signals, as in Decca. The locus of points of constant-range difference from two fixed points is one of a family of hyperbolae of which the two fixed points are foci. Thus from a pair of fixed stations a position line can be found; from a second pair a second-position line can be found, and hence the point of intersection is the position sought.

In the pulse systems, cathode-ray-tube methods resembling radar ranging methods are used for timing. With Decca, continuously rotating phase meters record the changes.

Gee operates in the band 20–80 Mc/s., Loran at about 2 Mc/s., and Decca in the neighbourhood of 100 kc/s. The operating ranges are about 100 miles, 600 miles, and 250 miles respectively, by daylight, using ground-waves. Sky-waves at night do not affect Gee, double the range of Loran, and halve the range of Decca. (Loran can discriminate and use sky-waves; Decca cannot discriminate sky-waves from ground-waves, with consequent errors.)

Low-frequency Loran on 180 kc/s. has had experimental trial and gives ranges of 1,500 and 3,000 miles by day and night respectively.

Decca has very high accuracy (0.05%), but can at present register only change of position. The pulse systems give absolute position, but with an accuracy not better than 0.3%.

Other Systems.—Consol, a system developed by the Germans, gives a bearing from fixed stations with an accuracy of 0.2 to 1 degree, to a range of about 1,500 miles, and uses a frequency of 200–500 kc/s.

Development effort in England and elsewhere is now being directed to giving useful radio aid to small ships at minimum cost. One approach is to attempt to provide by radio methods approximately the same facilities as those provided visually by a lighthouse. Proposed systems include (1) a beam which carries voice-modulation, speaking the bearing as it rotates; (2) two beams rotating, one at twice the speed of the other, and timed by a stopwatch to obtain bearing; and (3) a beam carrying pairs of pulses whose spacing varies with the bearing.

Radar Display Circuits and Techniques.

The principles of design of display circuits for radar search sets were discussed. Apparatus made in the Dominion Physical Laboratory and its use were described.