We captured new animals in spring to replace animals that died or were not pregnant with the goal to maintain 20 individuals fitted with VITs in each study area based on logistics and funding limitations. For the initial capture of female elk, we identified random points using a geographic information system GIS; ArcView, Environmental Systems Research Institute, Redlands, CA, USA within defined winter range boundaries and began our search at those points for the nearest group of adult females.
During autumn captures, we determined lactation status from presence of milk in the udder indicating a female had nursed a juvenile within the past 11 days Flook , Fleet and Peaker , Noble and Hurley If we could not express milk or if clear fluid was expressed, we classified the female as not lactating. We estimated nutritional condition of females using 1 a chest girth circumference to estimate body mass Cook et al. Two highly trained and experienced individuals R.
Mrs. Bradley Mysteries | Awards | LibraryThing
We inserted VITs Advanced Telemetry Systems into pregnant females during spring captures to aid in the capture of neonates and to pair neonates with adults of known nutritional condition. Methods for inserting VITs were described in detail by Johnson et al. Harvested animals provided uteri to determine pregnancy status, mammary tissue for lactation status, and lower jaw for aging.
Antlerless elk hunts were conducted the week of Thanksgiving Day just prior to the start of our late autumn captures. We used 3 primary methods to capture neonates: 1 monitored females with VITs to locate and capture neonates VIT neonate ; 2 observed behavior of solitary females from the ground or helicopter to locate and capture neonates by hand hand neonate ; and 3 searched by helicopter for females accompanied by neonates captured via net gun shot from the helicopter netted neonate. We followed protocols outlined by Rearden et al.
Horoscope and chart of Frank Robinson
Birth sites were characterized by the presence of a disturbed area about 0. Placental material, bodily fluids, or moist ground that attracted flies were also common at birth sites. If unsuccessful the second day, we usually returned a third morning to attempt to locate the neonate. If the VIT was not at a birth site, we assumed it was shed prematurely and did not search for the neonate. Once we observed a female tending a neonate, 1 or 2 people then walked to the site and captured the neonate by hand.
We opportunistically captured neonates by hand that we observed crossing roads, meadows, and other openings. When we located a solitary female with a bedded neonate during helicopter flights, the pilot landed in the vicinity and 1 person exited the helicopter and approached the neonate while the helicopter hovered over the neonate to distract it while it was captured we also refer to these as hand neonates. We conducted searches for neonates from a helicopter 3—8 days each year between 26 May and 12 June, timing the flights to occur just at or after peak parturition.
Total flight time for searches was 19—25 hours each year at both the SW and NE study areas except in when we conducted 45 hours of flights in NE. We conducted flights from approximately to and to in known parturition areas. When we captured a neonate, we blindfolded and restrained it, and determined its sex, weight to the nearest 0. For the survival analysis, if we could not determine a neonate's weight or sex because it was killed and partially consumed before we captured it, we randomly assigned a sex and the mean birth mass of neonates of that sex from that study area and year White et al.
We included these individuals in survival analyses because censoring these individuals would have positively biased survival estimates. To remedy that problem, we molded and constructed our own expandable collars Keister et al. As a final step, we glued 2. After being deployed a few weeks, the foam wore off the collar. Most juvenile mortality occurred the first few months of life, so the intensive monitoring schedule early in life allowed us to quickly locate dead juveniles and accurately assign a cause of death.
Flights usually started between and in summer and by in winter and continued until all collars were monitored. When a VHF signal from a collar indicated mortality mode, we recorded its approximate geographic coordinates with a GPS unit and ground crews then located the radio, investigated the site, and conducted field necropsy of the carcass. When we found a carcass that was partially consumed or with signs of predation, we skinned it, if necessary, and examined evidence of hemorrhaging, claw and bite marks, and crushed and broken bones to ascertain if the animal was killed rather than scavenged.
We looked for hair, scat, tracks, claw marks, drag marks, and bed sites of predators and disturbed vegetation, blood, and other sign of predation. When possible, we collected intact femurs from dead juveniles to estimate marrow fat percent Cook et al. We entered geographic coordinates for birth, death, capture, and relocation sites into a GIS database ArcView.
We did not consider individual precipitation covariates in combination with others given the high correlation among estimates. We reasoned that more severe winters would be reflected by years with greater precipitation and colder temperatures, resulting in increased snow depth. Thus, WSI increased in value as winter severity increased.
We then calculated a weighted mean NDVI across the study areas to account for different proportions of each open land cover type in the study area. We reasoned that greenness could fluctuate annually as a result of the cumulative precipitation from the previous 2 months to affect the nutritional status of females as they entered winter by providing more nutritious forage later into autumn. We assumed unmarked cougars killed within a study area met our requirements for inclusion. In this analysis, we were interested in identifying only cougars that were within our study area and not the size of their territories.
We continued to monitor hunting and damage reports through in NE and in SW to ascertain if any additional unmarked animals were killed in our study areas for inclusion in population reconstruction. We excluded subadult males from population estimates because they typically disperse from natal areas or are transient Logan et al.
We used mean IFBF spring for each year and study area as our response variable i. For these elk, we calculated mean IFBF autumn for each study area and year to use as our response variable i. We investigated whether nutritional condition of adult females explained variation in neonate birth weight or date using normal linear models R version 3.
For these analyses, we used either individual birth weight or birth date of juvenile elk as the response variable, and nutritional condition of its mother in autumn or spring as the explanatory variable. We identified relationships between capture mass and age at capture by sex for all VIT and hand neonates from NE using normal linear regression models. In all analyses, the individual was the sampling unit.
Dame Bradley books by Gladys Mitchell – a checklist
We developed a set of a priori candidate models for analysis, which were based on biological hypotheses, to test for potential effects of a multitude of variables on juvenile survival rates Appendix D through the first 30 days, 16 weeks, and 12 months of life. If cause of death was related to capture and occurred within 7 days following capture, we censored the animal from the analysis.
We followed the conventions of Lebreton et al. Alternately, in our most simple model, we hypothesized that survival was constant for all years, areas, sexes, and time periods: S. We also hypothesized that temporal variation in S would be constant. Trend models allowed us to determine if survival varied in a predictable manner as juveniles aged. Our modeling proceeded in 3 steps. First, we investigated models that accounted for effects of year, sex, and area on juvenile survival.
During this step, we constrained all models to a within year t effect on survival while we investigated combinations of year, sex, and area. Finally, we retained the best model that contained year, sex, area, and within year time effects and then added each covariate CoV i singly to identify covariates that explained variation in survival. We retained models when adding covariates that decreased AIC c scores below the model that did not include covariates i. During each step of our analysis, we used AIC c and w i for model selection and ranking Burnham and Anderson Because elk were single neonates no twinning and captured with different methods over large areas and at different times, we had no reason to suspect that there was any association i.
Consequently, we did not adjust for overdispersion due to lack of independence or apply a variance inflation factor. We used regression coefficients and their confidence intervals to evaluate strength of evidence of an effect for various covariates in models. To identify effects of maternal condition on juvenile survival, we restricted the analysis to VIT neonates.
We hypothesized that if juvenile survival was influenced by nutritional limitations operating at an individual animal level, we would see significant, positive effects of birth weight, birth date, or maternal nutritional attributes IFBF spring , BM spring , and age. In this analysis, we did not examine climatic covariates, which we used to represent a surrogate of nutritional condition, because we had direct measures of maternal condition.
Covariates considered in the analysis were sex, ordinal birth date, birth mass NE only , and age at capture of the neonate, cougar density, and 6 combinations of cumulative precipitation totals preceding parturition, WSI, and NDVI in October for each study area.
We estimated survival on an annual interval Jan—Dec and entered individuals into the analysis the first year they were monitored at the start of the interval e. The first was with all sources of mortality and the second where individuals that were killed by hunters were right censored in the year of death. For each analysis, we used a hierarchical modeling process.
After determining the best model that included temporal or study area effects, we modeled annual cougar density as a study area covariate or female age as an individual covariate to evaluate their influence on adult survival. We could not include nutritional condition of females as a covariate because we did not capture each elk every year that we monitored their survival. We completed 4 post hoc analyses to provide a better understanding of our results.
If capture had an adverse effect on juvenile survival, we expected the percent of females lactating in autumn to be lower for the cohort whose neonate we captured. We assumed that all females that successfully raised a juvenile through late November were still nursing; although, we found that 3 of 67 4. Our second post hoc analysis evaluated maternal experience to help interpret survival covariates.
- Export-Oriented Industrialisation: The ASEAN Experience (Routledge Library Editions: Development).
- A Course in Miracles: Original Edition Text - Pocket.
- Death Comes at Christmas.
- Comprehensive Treatment of Chronic Pain by Medical, Interventional, and Integrative Approaches: The AMERICAN ACADEMY OF PAIN MEDICINE Textbook on Patient Management;
- #mrsbradley Explore the world of Instagram - ywocugusyniz.tk!
- Book of Yes: A Reinterpretation of the Ten Commandments for the New Millennium.
Third, we used the known fates option in Program MARK to fully explore the effects of winter severity and precipitation in year t on survival of juveniles from December—May. Fourth, we attempted to differentiate if predation on juvenile elk was compensatory or additive mortality. We used a hypothesis framework Appendix A to determine if cougar or black bear predation was a compensatory, additive, or partially additive source of mortality. Cougar densities estimated from population reconstruction varied annually within and among study areas and were greater in NE than SW Table 3.
Densities of independent subadult female and adult cougars in NE at the start of elk parturition 15 May were 2. In NE, cougar densities increased in Sled Springs during the course of the study, whereas in Wenaha they decreased from — but subsequently increased in and Variation in cougar density in NE was primarily attributable to variable annual hunter harvest. From 15 May through 14 May , no cougars were killed in either study area in contrast to — when 9 cougars were killed in Wenaha or — year prior to start of capture work when 10 cougars were killed in Sled Springs.
After both years of high hunter harvest, estimated cougar densities were low the following year and then subsequently increased. In SW, cougar densities were relatively stable at Toketee but increased at Steamboat during — Densities of independent female and adult cougars were 0.
- Virginia Woolf: The Will to Create as a Woman!
- Navigation menu.
- Peptide Information 3817?
- Publication Order of Mrs. Bradley Books?
- The Writer!
- BDU: Bootcamp Diary Unauthorized?
- Surviving Mom!
- Effective Substance Abuse Detox.
- Claudine At School (Vintage Classics).
- The Enchanted Screen: The Unknown History of Fairy-Tale Films.
No unmarked cougars were reported killed within Toketee and Steamboat study areas during or in the 4 years following data collection. He remained in the study area and included in the population estimate. Although cougar harvest was less in SW than NE, this low level of mortality from hunting combined with increased natural mortality rates in this region Clark et al.
We identified 32 individual bears 11 females, 21 males in Sled Springs and 49 individual bears 19 females, 30 males in Wenaha from DNA analysis of hair. Within study areas, estimates of bear populations in Sled Springs or Wenaha were similar for models that included heterogeneity or time, but models that accounted for behavior produced estimates that were larger with broad confidence intervals; consequently, we dropped behavioral models from our population estimates.
These estimates indicated black bear densities were relatively similar among study areas and years. We captured 4 yearling and adult female elk times across 4 study areas with and capture events in spring and autumn, respectively. In Sled Springs, we captured 68 females times in spring and times in autumn. In Wenaha, we captured 53 females times in spring and 67 times in autumn. In Steamboat, we captured 31 females 54 times in spring and 25 times in autumn. In Toketee, we captured 44 females 61 times in spring and 24 times in autumn.
Ages of females that received VITs generally increased over time within a study area because our study design relied on repeated measurements of nutritional condition of females Figs. During and when we caught VIT neonates in all 4 study areas, ages of pregnant females did not differ among study areas or between years Table 4. In autumn, measures of nutritional condition of BM autumn and IFBF autumn of females 2—14 years old varied significantly by pregnancy and lactation status. For lactating elk, neither age nor BM autumn differed among study areas. Among study areas, however, IFBF autumn of lactating elk differed and was 2.
Mass of females did not differ among the other 3 study areas Appendix H; Fig. Mean age of females in NE observed with delayed breeding These results suggest that mean nutritional condition entering winter had a much stronger effect on mean IFBF spring than any of our measures of winter weather in NE. There was no evidence to indicate cougar density influenced female survival because this model ranked lower than the constant survival model. Overall, annual survival including all sources of mortality was 0. We identified an additional 56 birth sites in NE where we did not capture the neonate.
Mean ages at capture were 1. Given that timing of conception was influenced by nutritional condition Cook et al. From —, we captured neonates in NE: from elk with VITs, from behavioral observations of solitary unmarked elk, and using a net gun fired from a helicopter. Median birth date for hand and VIT neonates was day We captured 2 neonates 1 from a female with a VIT that were abandoned and in poor condition when we found them and subsequently died of starvation. One VIT neonate with a deformed hoof was killed by a black bear 3 days after its capture.
Even though skin wounds had healed when carcasses were recovered, we are unsure the role, if any, the injuries played in their deaths. We collected 21 femurs from juveniles killed by cougar or unknown predation from 1 December through the end of May. We captured neonates in — 46 from VITs, 42 from behavioral observations of solitary unmarked female elk, and 68 using a net gun fired from a helicopter. The VIT neonate captured on 28 April weighed 6.
We judged it to be born prematurely and censored it from survival analyses. Unknown predation included 8 probable cougar and 2 probable bear predations. We did not collect femurs from mortalities in SW. We assumed lactation status predicted juvenile survival through late November. Female mass and IFBF spring were highly correlated, explaining why both appeared as competing models and were never included in the same model. When we included time trend effects i. Three models were within 2 AIC c points of the top model.
Competing models included covariates for female age and cougar density Appendix L. Time trend model LnT was 3. Model selection results for the analysis of VIT neonates to 12 months followed similar patterns for survival to 16 weeks. Two models were competing; in one model, female age entered the model and in the other model region replaced cougar density Appendix M.
In models where region replaced cougar density, survival was higher in SW where cougar densities were lower. There was no support for the other covariates. Survival rates differed by study area, but a pattern of increased survival was consistent over time during the first 30 days, 16 weeks, and 12 months of life. The negative relationships of juvenile survival with female mass or IFBF were contrary to our initial hypotheses.
In a post hoc analysis attempting to explain the negative relationship of IFBF and mass with juvenile survival, we found evidence that maternal success varied among females. Because lactating females were thinner in both autumn and spring across all study areas, this could explain the negative relationship between juvenile survival and maternal condition we observed. Survival was higher when cougar density was lower Fig. Signs of coefficients were positive for capture age and sex male but negative for birth date and cougar density Fig. Competing models included effects of cougar density and birth date Appendix P.
Signs of coefficients were positive for capture age and negative for cougar density and birth date Fig. Consequently, we were unable to identify any abiotic or biological factors that substantially influenced juvenile survival from 7—12 months of age and this indicates variation in environmental factors was only weakly associated with juvenile survival during this period in NE. In contrast to NE, the best performing models for survival of juvenile elk in SW included abiotic factors and cougar density was rather unimportant in explaining survival of juveniles.
April—May total precipitation entered into the models with a negative coefficient Fig. Birth date entered into the top model with a positive coefficient Fig. December—February precipitation t — 1 entered into 1 competing model with a positive coefficient Appendix R. Across our 4 study areas, survival of juveniles was strongly related to cougar density. We found that variation in cougar density explained the majority of variation in juvenile survival through the end of summer and recruitment of juvenile elk into the adult populations.
We observed minimal effects of climatic variation on vital rates of elk, but some factors e. Although these results were consistent with other studies that demonstrated the strong effect of predation on juvenile elk survival Myers et al. Population growth rates of elk tend to be most sensitive to adult female survival followed by juvenile survival, and finally pregnancy rates Raithel et al. Harvest of adult female elk has the greatest effect on elk population dynamics Clark , Eacker et al. As wildlife managers reduced antlerless elk hunting to maintain elk populations Brodie et al.
Using empirical data observed from field studies conducted in northeastern Oregon e. Pregnancy rates of female elk were influenced annually by lactation status, elk density, August precipitation, and previous year winter severity see results from Johnson et al. At each step of the simulation, values of covariates known to influence pregnancy e. Using these randomly generated covariate values, estimates of pregnancy and survival were calculated based on observed relationships to populate the Leslie matrix at each time step of the simulation.
Other parameters in the Leslie matrix e. At the conclusion of simulations, Clark regressed effects of covariates against estimated population growth rates to determine the relative effect, measured by the slope and fit of the regression, for each covariate and vital rate on elk population growth rates. Similar to Eacker et al. However, in cases with lower cougar densities, elk populations would still increase under mean environmental conditions. The results of Raithel et al. We were unable, however, to detect negative effects of increased winter precipitation or severity on spring body fat.
Summer drought can decrease forage quantity and quality Vavra and Phillips , Weisberg et al. These effects suggest that for NE, summer precipitation indirectly influenced elk population performance. The relatively mild winters in our study areas suggest that effects of winter weather on juvenile survival may be less severe compared to other elk populations in the northern Rocky Mountains.
For survival analysis of all neonates captured in NE, climatic covariates were uninformative predictors. In contrast, in SW several climatic covariates explained variation in juvenile survival through either 30 days or 16 weeks but not to 12 months. We hypothesize that this may be due to wetter conditions in early spring that resulted in increased vulnerability associated with hypothermia. Remotely sensed metrics, such as NDVI, may indirectly represent available forage with higher levels of greenness and forage quality Pettorelli et al.
The negative influence of December—February precipitation totals suggests harder winters most precipitation during this time is likely snow will decrease overwinter survival of juvenile elk, which was expected. During periodic severe winters, or if winters become increasingly severe in the face of climate change, overwinter survival of juvenile elk would likely decline Proffitt et al.
We measured and tracked adult female nutritional condition and reproductive performance in early spring and late autumn and linked adult condition to reproductive status, neonate birth mass, and survival. In general, IFBF autumn of lactating females is a reflection of the balance between energy expenditure and the nutritional value of summer and early autumn forage Cook et al.
Using these criteria, we documented 3 lines of evidence of nutritional limitations in our study areas. In SW, females had lower body fat than many populations in the northern Cascades but higher than observed in coastal environments Cook et al. The IFBF autumn levels were a third to a half of what elk are able to achieve if provided excellent nutrition throughout summer and early autumn see Table 9 in Cook et al.
Cook et al. This spatial pattern was evident in nutritional condition of wild elk in the Cascades with elk in northern environments having higher body fat Cook et al. This suggests variation in regional nutritional resources is associated with broad regional climatic factors, ecological context, disturbance, and succession that have substantial influences on nutritional condition and reproductive performance of elk Cook et al. By not capturing females and assessing pregnancy until early December, our approach to document delayed breeding could only identify those elk breeding at least 1.
Perhaps more telling, we found an inverse relation between IFBF autumn and birth date the following spring. Even when pregnancy rates are high, if nutrition is limiting, ovulation may be delayed; the higher the nutritional limitation, the later the breeding date Trainer , Albon et al. This effect of inadequate summer and autumn nutrition is significant because timing and synchrony of birth date may be important for survival by allowing juveniles to increase body mass and be able to better withstand extreme winter events Keech et al. Penguin gifts. Writing workshops. View all. Events Podcasts Apps.
Contact us Contact us Offices Media contacts Catalogues. Home Gladys Mitchell Gladys Mitchell. Books Biography. Pre-Order From. Murder in the Snow Gladys Mitchell. Watson's Choice Gladys Mitchell. Death and the Maiden Gladys Mitchell. Death at the Opera Gladys Mitchell. Spotted Hemlock Gladys Mitchell. Nest Of Vipers Gladys Mitchell. Dead Men's Morris Gladys Mitchell. Brazen Tongue Gladys Mitchell. St Peter's Finger Gladys Mitchell. The Devil's Elbow Gladys Mitchell. For it was honest morning now, a September morning, blowing wild-grapes and sea sand and bayberry into Roger's nostrils.
Biography of Frank Robinson (excerpt)
As he stared at the house a great hound crept around the corner of it, baying monotonously, but as he saw Margarita he left off and ran to her, arching his brindled head. He was a Danish hound, beautifully brindled and very massive. She fondled him quietly, smiling as he clumsily threw his great paws about her waist, and pushed him down. She turned her direction slightly and made for the cow stall, and as he stood by the door Roger saw that whatever the internal structure of the building might be, it was certainly covered with rough sand.
At sight of Margarita his jaw dropped, he shivered violently and appeared ready to faint, but as she called encouragingly to him he mustered courage to approach and feel of her skirt timidly. He was evidently feeble-minded as well as dumb, for with a sort of croak he dropped the bucket and began to dance clumsily up and down, snapping his fingers the while.
Plainly he had thought her gone for good and this was his thanksgiving. Make haste with it. He started on a run for the door furthest from the cow stall and appeared almost immediately with a large silver mug and a huge piece torn from a loaf. Squatting beside the cow he balanced the mug between his knees and deftly milked it full. She seized it, drained it thirstily and began munching her bread, holding the mug out to him again to be filled a second time.
She bit great mouthfuls from the loaf, like a child of four, and Roger watched her, half amused, half irritated. She shook her head doubtfully. I should not like it. If Caliban will get you another He did that when the doctor came to see my father. I asked Margarita a year or two after this to describe for me how she first entertained Roger: I had already a good idea of his initial hospitality to her in the French restaurant.
Here is her letter. What an odd thing to ask me to tell you—my first hospitality to Roger! But I remember it very well. Only it was not very hospitable, because, of course, I did not know anything about that sort of thing. One has to learn that, like finger bowls and asking people if they slept well. You know I called for some bread and milk and ate them very greedily, standing by the cow so that I could get more when I should want it.
By the time I had finished, Caliban had finished milking and then Roger asked me quite politely if I thought he might have something to eat now. You know, dear Jerry, I had never been used to eating with people. All the people I knew ate their meals separately and it never occurred to me that I ought to be there when he ate.
And then, I was so sleepy—oh, so sleepy! You know I have always felt sleepy and hungry and angry and things like that so much more than other people seem to. I have to sleep and eat when I feel like sleeping and eating. So I only said, "You had better ask Hester to get you a breakfast. I must go to sleep now," and flung myself down on some fresh hay just beside the cow stall, in the sun, and went to sleep! Was not that a dreadful thing to do? But I did it. I do not know how long I slept, nor how Roger looked when I turned my back on him, but when I opened  my eyes he was sitting beside me, smoking a cigar and staring at me.
He had been there all the time. So we went in, but Hester was not in the kitchen, and when I went up to her room and knocked there was no answer, so I supposed she had gone out for the roots and herbs she used to hunt so much. I think now it was because I began to understand that I ought to have done something I had not. Why do you not ask Caliban? So then he asked Caliban if he could manage some breakfast for him, but Caliban only stared and walked away. I am sure he knew it was not that I grudged him food, but that I had no idea at all of how to set about getting it ready.
People always have known that what I say is truth, though much of what I say seems to surprise them. He lifted Caliban in the air by the collar of his coat and gave him several sharp blows on each ear and shook him. Then he threw him away on the floor. Caliban cried like a young dog and sat upon his knees and covered his face. He meant  for Roger to excuse him. I was surprised, for I had always been a little afraid of Caliban. And see that you obey me in future. Caliban hurried about and looked here and there and made some coffee and broke eggs in a black pan and cut pieces of bacon.
He set a place at the kitchen table and made some biscuits warm in the oven. Roger ate five eggs and a great many pieces of bacon and six biscuits. He gave me some coffee. When he had finished he drew a long breath and gave Caliban a piece of silver money and Caliban kissed it. Then Roger took another cigar and told Caliban to fetch a match and then he asked me if I would like to walk by the sea for a little.
I am too comfortable. Will you come out with me? So I said I would, and that was all my hospitality, dear Jerry.
I had learned better when you came, had I not? This letter has been so long that I cannot write any more. My Margarita! The very words are not like any other two words. I think no woman's name is so purely sweet to the ear, so grateful on the tongue. Alas, alas As to that walk by the sea, I have never been able to get any satisfactory account of it. Any, that is, which could hope to prove satisfactory to one who did not know Roger. Such an one might be incredulous, in face of all that had gone before, when assured that Roger paced back and forth on the firm sand, filling his lungs in the clean sea air, puffing his cigar in perfect silence, Margarita at his heels as silent as he, and the big Danish hound at hers, more silent than either.
But so it was. To me who know them both, nothing could seem more natural. They were healthy, well-poised animals, well fed, supplied with plenty of fresh air  a prime necessity to them both and in congenial company. Neither of them was given to consideration of the past or prognostication of the future; both of them were content. Roger has always had that priceless faculty of reserving mental processes, apparently, until they are necessary. When they are not, he lays them by, as a sportsman lays by his gun, and the teasing, relentless imps that poison the rest of us with futile regrets for the past and vain hopes for the future avoid him utterly.
It is the pure Anglo Saxon corner-stone of that great, slow wall which I firmly believe is destined to encircle the world, one day. Your slender, brown peoples with their throbbing, restless brains and curious, trembling fingers may—and doubtless will—build the cathedrals and paint the frescoes therein and write the songs to be sung there; but they must hold their land from Roger and his kind and look to him to guard them safe and unmolested there.
Or so it seems to me. After an hour or so of this walking Caliban approached them, and bending humbly before Roger made it clear that he greatly desired their presence at the cottage. They went after him, Margarita incurious because she was utterly indifferent, Roger wasting no energy, of course, with no facts to proceed upon. At the kitchen he endeavoured to lead them up the narrow stair, and then Margarita asked him if anything was wrong with Hester and if she had sent him.
She is sitting in her chair. Her eyes are open and she is dead. Roger stared thoughtfully ahead of him. He never doubted her for a moment. It was always impossible to doubt Margarita. This woman, if I understand you, has taken care of you from babyhood! She liked my father. It may seem strange to you that Roger did not immediately ascend the stair and confirm Margarita's report, but he did not. Instead he spoke to Caliban. The clumsy, slow-witted youth nodded his head and sobbed noisily, with strange animal-like grunts and gulps. Caliban raised his hand and checked off the five fingers slowly.
It was understood that he indicated so many hours. He placed his hand upon his heart, then shook his head from side to side. Suddenly he shifted his features unbelievably and Roger gazed horrified upon a very mask of death: there was no doubt as to what Caliban had seen. I think I will take a little nap on the beach, if you don't mind, and then I'll go to the village and get help to—to do the various things that must be done. Later I will have a talk with you.
Tell me once again—you do not know of any friends or relatives of your father's or Hester's? But this question was beyond the poor lout's intelligence; he could only blubber and fend off possible chastisement. Call me if you want me. She went off to her warm straw, threw herself on it like a tired child, and passed quickly into a deep sleep; he tramped for a moment on the beach, then stretched himself in the lee of a sun-warmed rock and fell into the dreamless, renewing rest that he took as his simple due from nature.
When he woke it was full sunset. The lonely reefs were red with it, O Margarita, well I know that hour! Do you remember our talks? In just that glow I have watched it, leaning on my oars, and for a few strange minutes, the exact time necessary for the sun to drop behind the coast-hills, I have felt myself a small boy again, crouched in a cane chair before my mother's sewing-table, unable for very terror to drop my feet to the floor as I gazed through wide eyes at the House of Usher, that home of sunset mystery.
Such a strange, Poe-like atmosphere could that sanded, secret cottage take upon itself. Roger pushed rapidly up the beach and entered the house quietly, so quietly that he caught Margarita's last sentences, which struck him as odd even in his utter ignorance of their connection. She was evidently scolding Caliban, for his grunts and shufflings punctuated her pauses. I know she can not walk any more. My father could not walk when he was dead. And you need not think that Roger Bradley will not ask, because he will.
He knows everything. Roger thought that the lout had been teasing her with stupid ghost hints and bade him begone sternly, more vexed than before as he noticed the dim twilight drawing in and  realised how late and inconvenient the hour was for all he had to do. Hester," she returned, "you cannot do anything to Hester, Roger Bradley, for she has gone. The room that she indicated faced the stairs directly. It was furnished plainly with an ugly wooden bed covered with a bright patchwork quilt, a pine bureau and two cheap chairs.
The walls were utterly bare and the floor, but for a woven rug near the bed, of the sort so common in New England. And yet there was an air of homely occupation in the plain chamber, a bright, patched cushion in one chair, a basket full of household mending and such matters, on a small table, a pair of spectacles and a worn Bible beside it.
The room had that unmistakable air of recent occupation, that subtle atmosphere of use and wont that no art can simulate—and yet it was empty. Roger came down the stairs again and summoned Caliban. The fellow lay in a deep sleep, just as he had thrown himself, on the straw beside the cow stall, a full pail of milk beside him. It was hard to wake him, for he scowled and snored and dropped heavily off again after each shaking, but at last he stood conscious before them and appeared to understand Roger's sharp questions well enough, though his only answer was a clumsy twist of his large head and a dismal negative sort of grunt.
Where was Hester's body? Was she really dead? Had anyone been in the house? What had he been doing all the afternoon? One might as well have asked the great hound in the doorway. Even to threats of violence he was dumb, cowering, it is true, but hopelessly and with no attempt to escape whatever penalty his obstinacy might incur. We shall have to have more bread and milk.
Let us eat it on the rocks, Roger Bradley, will you? And Roger, in spite of the fact that he was forty and a conspicuously practical person or was it, perhaps just because of this fact? I confess I am not quite sure! They sat on the rocks, warm yet with the September sun, and ate with a healthy relish, while the first pale stars came out and the incoming tide lapped the smooth beach. I have been assured that they never in the conversation that followed mentioned the island—though it was not then an island, to be sure—that they were sitting upon, nor the extraordinary events which had happened there and had brought them to it.
And I believe it. I also believe, and do not need to be assured, that they talked little of anything. They never did. Again and again I have imparted to Roger some or other of Margarita's amazing conversations with me and he has listened to them with the grave interest of a stranger and even questioned me indolently as to my theory of that stage of her development. I must add that he  has never seemed surprised at what she said and has occasionally corrected me in my analyses and prophecies with an acuteness that has astonished me, for he was never by way of being analytic, our Roger.
When I once remarked to Clarence King who was devoted to her apropos of this silence of theirs that it was like the quiet intimacy of the animals, he looked at me deeply for a moment, then added, "Or the angels, maybe? I never heard him in my life talk so brilliantly as he did one afternoon stretched on the sand by Margarita, while she fed him wild strawberries from her lap and embroidered the most beautiful butterfly on the lapel of his old velveteen jacket, and Roger tried to ride in on the breakers like the South Sea Islanders.
From time to time Clarence would turn one of those luminous sentences of his and kiss the stained finger tips that fed him I never did that in my life and from time to time Roger's splendid tanned body would rise between us and the sun, triumphant on his board or ignominiously flat between the great combers. But he was as calm as the tide and we knew that he would beat it in the end and "get the hang of it" as he promised.
She never turned her eyes toward him, that I could see, but I am convinced that she was perfectly aware each time he fell. She never talked much to King and he was always a little jealous of me on that account. But she was very fond of him and always wrote to him when he was off on his ramblings. His letters to her were always in rhyme, the cleverest possible. There are, of course, whole pages to be written—if one wanted to write them—of that night on the rocks.
I naturally don't want to write them. To say that I have not imagined them would be a stupid lie; I am human. But I have never been able to bring myself to the point of view of the modern lady novelist in these matters. Why is it, by the way, that God has hidden so many things in these latter days from the prudent and revealed them unto spinsters? Not that I need to rely on my imagination: Margarita would have saved me that. Once she got the idea that I was interested in those early days, she was perfectly willing to draw upon her extraordinary memory for all the details I could endure.
But of course I could not let her. The darling imbecile—could anything have been so hopelessly enchanting as Margarita? It is impossible. If you can picture to yourself a boy—but that is misleading, directly, when I think of her curled close against me on the rocks, her hand on my arm and all my veins tingling under it. She was all woman.
It was like those dear talks with some lovely, loved and loving child. But that, again, gives you no proper idea. For no child's throat sounds such deep, bell-like tones, such sweet, swooping cadences. And no child's eyes meet yours with that clear beam, only to soften and tremble and swim suddenly with such alluring tenderness that your heart shakes in you and slips out to drown contentedly in those slate-blue depths. No, no, there is no describing Margarita. Perhaps King came nearest to it when he said that she was Eve before the fall, plus a sense of humour!
But Eve is distinctly Miltonian to us unfortunately for the poor woman and Margarita would have horrified Milton—there is no doubt of it. Well, well, I left them on the moonlit rocks, and there I had better leave them, I suppose. It is so hard for me to make you understand that Roger was incapable of anything low, when I am apparently doing my best to catalogue actions that can be set only too easily in an extremely doubtful light.
All I can say is, pick out the best fellow you know, the one you'd rather have to count on, at a pinch, than another, the one you'd swear to for doing the straight thing and holding his tongue about it—then give him five feet eleven and a half inches and blue eyes and you've Roger. This is rather a poor dodge at character  drawing: I know a competent author would never throw himself on your mercy so. But then, what does it matter? When the members of a man's own household, who have known him from boyhood, fail to understand him and take a satiric pleasure in looking at what he does from the nastiest possible standpoint none the less nasty because it is a logically possible standpoint why should I, a confessed amateur, hope to make Roger clear to you if you are determined to misjudge him?
I find myself still a little sore on this point: unnecessarily so, you may be thinking. But you never had to explain it to the family in Boston, you see—and Sarah. I had. I can see her cold, grey-green eyes to this hour, her white starched shirt and her sharp steel belt buckle—ugh! It should be illegal, in a Republic where there are so many less sensible laws, for any woman to be so ostentatiously unattractive See how it has crept up and cut them off.
Do you not remember, I told you how he carried the blueberry pie and the milk out there and we ate them? He was so hungry! It was then that he looked at me so——". It makes one's mouth so black. And then he said that mine was worse, because there was some on my chin—why do you scowl so, Jerry? Is that a wrong thing to tell? How do you see, Jerry? But I was telling you about the tide, was I not? When Roger said that about my mouth I tried to get the  stain off, but I could not, and then Roger said it was no use trying any more and he kissed me.
Here Margarita paused and patted my hand, tapping each finger nail lightly with her own finger-tips. Margarita's repulsed fingers lay loosely upcurled on her knees, which she hunched in front of her, like a boy. It is easy to see that I should have made a poor novelist; it has been hard enough for me to give you any idea of scenes I did not myself witness, even though I had Roger and Margarita to help me out and an intimate knowledge of both of them, and when I try to fancy myself composing a tissue of fictitious events "all out of my head," as the children say, my pen drops weakly out of my fingers, in horror at the very thought.
But now, thank heaven, the pull is over. From now on, I need tell only what I knew and saw, in the strange, interwoven life we three have led. Three only? Nay, Harriet of the true heart, Harriet of the tender hand, could we have been three without you? My fingers should wither before they left your name unwritten.
I remember so well the night the telegram came. I had been vexed all day. Everything had gone wrong. Roger, to meet whom I had come back early to town, had neither turned up nor sent me any message; the day had been sickeningly hot, with that mid-September heat that comes to the Eastern States after the first crisp days and wilts everything and everybody. I found my rooms atrociously stale and dusty, and worse than that, perfectly useless, since by some miracle of carelessness I had left my keys behind me at the shore and hadn't so much as a clean collar to look forward to.
The club valet assured me that he had received no call for trunk or bag, but that Roger had assuredly not entered the house for five days. I went into his rooms, but they told me nothing, and I, worse luck, should have been lost in his  collar, so I glared angrily at the drawers of linen, wired for my own keys and made for the Turkish bath. There with a thrill of delight I discovered a complete change of clothing; I had, before leaving for the summer, jumped hastily into dinner things, leaving a heap of forgotten garments behind me and they awaited me now, trim and creased, russet shoes polished, and a wine-colored tie, a particular favourite of mine, topping the fresh linen.
It seems absurd, but I recall few moments in my life of such pure, heartfelt thanksgiving. The very colour of life seemed changed for me. I wonder if we do well in despising these small thrills as we do? Surely enough of them sedulously preserved in grateful memory must equal in intensity those great, theoretical moments we all regard as our due but so often pass through life, I am sure, without experiencing. However that may be, the little gratifications of that evening are graven in my mind, undoubtedly, you will say, because of the startling climax for which they were preparing me.
The clean tingling of my soapy scrub, the delicious coolness of the plunge, the leisurely, fresh dressing all caressed my nerves delightfully. In the plunge a pleasant enough fellow had accosted me and we had splashed together contentedly. I expected to recall his name every moment, for his face was vaguely familiar, but I could not, and when we met in the hall and went down the steps together, it still escaped me.
We hesitated a bit on the pavement, and then before I realised it we were hailing a hansom and bound for dinner together. It was a pleasant drive up along the river, for a little breeze had sprung up and the watered asphalt smelt cool. We were both comfortably hungry and very placid after our bath and we chatted in a desultory sort of way, I, amused at my utter inability to place the fellow, he quite unconscious, of course, and perfectly certain of me. He asked after Roger, sympathised with our failure to make connections, remarked to my surprise that he had only been out of town for his  Sundays America had not adopted the "week-end" at that time and asked me, I remember, if I knew anything about a game called basket-ball.
It seemed he was anxious to find someone who did. We drew up at last to our white, glistening little table looking out over the water, looked about for possible friends, nodded to the head-waiter and ordered our dinner. It turned out that neither of us had yet celebrated the oyster month, and leaving my unknown to bespeak the blue points, for the more conservative among us clung to the smaller oyster then, I telephoned the club to let Roger know where to find me in case he should appear there.
Over the soup my companion got on to the subject—somehow—of evolution, and talked about it very ably indeed. I remember that those quaint and apparently highly important beasts lasted well into our guinea-chick and lettuce-hearts, and I can see him now, his eager, dark face all lighted with enthusiasm while he spread mayonnaise neatly over the crimson quarters of tomato on his plate, and made short nervous mouthfuls, in order to talk the better. Half amused, half interested I listened, trying to place the fellow, but for the life of me I could not.
Was he a scientist, a lecturer, a magazine writer, a schoolmaster? We finished with some Port du Salut and Bar-le-duc—an admitted weakness of mine—and I had decided to regularly pump him and find out his name without his guessing my game, when he began as I supposed, to help me out. But you'll remember, perhaps, this used to be a sort of hobby of mine, and I work it into shape nowadays for a young men's club I'm running. I yawned and lit a cigar and we sipped our coffee in silence. We think, when we are young, that we live alone. I recall, as a boy of twenty, certain hot-headed, despairing midnight walks when the horror of my hopeless, unapproachable, unreachable identity surged over me in melancholy waves.
I would have plunged into a monastery if I had believed that any sort of prayer and fasting could bring me close—really close—to God; for to any human creature, I had learned, I could never be close. After that, we grow into that curious stage of irresponsibility which we deduce from this loneliness, and distress our patient relatives with windy explanations of "matters that concern ourselves alone.
I like to believe, now, that a dim idea of what was going to happen was in some mysterious way growing on me before I got the telegram. I am certain that when the head-waiter touched my arm and told me I was wanted at the telephone, a curious oppression fell over my hitherto contented after-dinner spirit which grew into a kind of excitement as I made my way to the booth. And yet I expected nothing more than to hear Roger's voice with some reasonable explanation of his failure to meet me. It was the night porter, however, reading me a telegram missent to the shore and returned to the club.
He mumbled the name of a place I had never heard of and went on in the peculiarly expressionless style consecrated to messages, thus transmitted. Tell Hodgson to pack a complete change for Mr. Bradley and his razors. And see if you can find the name of the place from the chief operator and the correct message. It can't be parson, of course. And look up the next train for that place, if you can, Richard. I'll be down there directly.
I puffed hard at my dying cigar and went slowly back to the veranda, trying to make sense of that telegram. He wants me to meet him at some place or other at present unknown, and to bring him his razors and a sensible parson. Any special denomination? Anyhow I must start directly. There may be a night train. Would you rather stop here a while? Perhaps he's going to fight a duel with the razors and wants the parson for the other fellow!
Perhaps he's made a bet to shave a parson. But I was in no mood for joking. The telegram, so unlike Roger, and yet so unmistakably his, in a way—I have often noted a curious characteristic quality in telegrams—worried me. I wished I had got it in time to make the train he mentioned. I wished I were in that mysterious town. Suppose he had depended on me for it? Suppose he needed me? We drove down in silence. My man got out with me at the club and smiled at the Gladstone the porter held out to me.
Richard had the name of the town for me, too the town I prefer not to tell you and the next train that would make it: it left in fifteen minutes. Shall I call you a cab, sir? In the first place, I don't believe there is such a thing! The sooner we get through with all this white choker and black coat business, the sooner we'll amount to something, in my way of thinking. Well, seriously—will I do? Do you know anybody better? Because I'll go, if you don't. We'll have to hurry, I'm afraid. We had just time to jump for the last platform.
I remember apostrophising the Gladstone rather strongly as I fell on its metal clasp, and glancing apologetically at my companion, but he was tactfully deaf, and we found a seat together, by good luck, and settled down for our hot and tiresome night. I couldn't very well ask his name by that time, it would have been too absurd.
I trusted to Roger to get me out of that difficulty, for he knew Roger, evidently, and me too, though not very well, I judged. He certainly wasn't in my college class, for it would have come up, I was sure, in our talk. Not that we talked much. It was a stuffy, disagreeable ride, and I was alternately vexed with Roger and worried about him.